The Role of Social Media in Developing EFL Learners’ Speaking Fluency, Pronunciation, and Confidence in an Exam-Oriented Context

Zhang Xinyu1*, Chandra Mohan Vasudeva Panicker2

1&2 School of Education, Lincoln University College, Malaysia

^*Corresponding author’s e-mail: xinyu.masterscholar@lincoln.edu.my

Received: 21 Sep 2025

Revised: 01 Jun 2026

Accepted: 06 Jul 2026

DOI: doi.org/10.60072/ijeissah.2026.v4i03.001

ABSTRACT

In exam-oriented schools, students often lack the chance to practice speaking English in a meaningful way; however, social media serves as a pivotal digital ecosystem where learners autonomously interact with the language. This study investigates how social media helps students develop their speaking skills in these exam-focused settings, with a focus on fluency, pronunciation, confidence, and exposure to diverse English styles. Using a mixed-methods approach, the study collected data through questionnaires and semi-structured interviews to explore the experiences of EFL students in Guangxi, China. The findings show that social media supports personalized learning by allowing students to practice at their own pace and engage with content they prefer. In particular, participants reported higher confidence and better pronunciation through repeated exposure and the imitation of various English accents. Ultimately, the study concludes that social media acts as a helpful resource that fills the gap left by classroom teaching, offering context-specific strategies to bridge the digital usage gap between urban and rural learners.

Keywords: EFL Speaking Development; Exam-Oriented Education; Mixed-Methods Study; Personalized Learning; Social Media-Mediated Learning

Published by Asia Pacific Higher Learning Sdn Bhd., Malaysia

Background

English is largely a test preparation subject in school systems that emphasize exams. As a result, reading, grammar, and writing are typically the focus of teaching. While these skills are important for grades, students often have very few chances to engage in meaningful spoken interaction in class. Concurrently, digital technology has established new paradigms for extramural learning; global platforms such as TikTok, Instagram, and YouTube have emerged as critical ecosystems for autonomous language acquisition (Alsharari & Mohammed, 2026; Razali, 2021). Within these digital 'wilds,' students can internalize authentic native pronunciation models in a low-anxiety environment (Liu & Darvin, 2024). Accordingly, this study examines the pedagogical potential of social media platforms as compensatory tools within exam-driven secondary education (Udam, & Mohammed, 2023).

In the context of this study, exam-oriented education is defined as an instructional system where teaching methodologies and curriculum design are strictly driven by standardized testing requirements, often at the expense of communicative language practice. The Gaokao strictly directs English education in China, forcing classrooms to focus heavily on reading and grammar. Consequently, oral proficiency development is frequently marginalized in instructional time allocation. This leads many students to achieve high written scores while still struggling with actual conversations. Many learners respond by going online for the practice they miss in class. They frequently use apps like TikTok and YouTube and rely on pop culture to hear real-world English. As Wardani (2023) and Mu’aggil & Pido (2023) observe, this move to digital platforms is constantly growing. Following this trend, this research examines how students use these tools to develop their oral skills away from exam-focused instruction (Wardani, 2023).

Objectives

A key concern is that speaking practice remains secondary in exam-oriented high schools. Since the Gaokao primarily tests reading and grammar, teachers typically allocate classroom time to test preparation. Wardani (2023) notes that digital apps support independent practice. However, such studies frequently focus on university students with flexible schedules. This reveals a clear gap in the existing literature: it remains unclear whether these digital tools effectively help teenagers who operate under intensive exam demands. Consequently, as Liu (2025) explains, the efficacy of mobile applications is highly mediated by geographical and socioeconomic factors, precluding a one-size-fits-all solution. To address these realities, this study explicitly aims to investigate how secondary learners in Guangxi utilize platforms such as TikTok, Instagram, and YouTube to enhance oral proficiency, evaluating the efficacy of these digital ecosystems in mitigating formal instructional gaps. Consequently, this research focuses on two main questions:

How does social media influence the speaking skills of English learners in exam-focused high schools?
In what ways does social media act as a tool to fill the gaps left by classroom instruction?

Literature Review

This study is grounded in theories of second language acquisition and situated within the realities of exam-oriented English education in China. The review first examines how traditional teacher-centered approaches, particularly grammar-translation practices shaped by high-stakes examinations, influence speaking development. It then discusses relevant SLA theories—including input-based and interaction-based perspectives—that help explain how social media may support informal oral practice. Finally, the section considers personalized digital learning and identifies gaps in research concerning speaking development within exam-driven secondary education contexts (Alsharari, & Mohammed, 2026; Udam, & Mohammed, 2023).

Exam-Oriented Education and Speaking Development

The demands of the Gaokao heavily shape secondary English classes. Teachers often rely on the Grammar-Translation Method just to manage the strict syllabus. The result is that students tend to view English as a set of rules rather than a real communicative tool. Since speaking is excluded from high-stakes testing, teachers allocate almost all class time to written exercises. Viewed through the lens of Krashen’s (1985) Input Hypothesis, this exam-oriented approach clearly restricts comprehensible input. The high pressure also increases the affective filter (Krashen, 1985), meaning students actively avoid oral practice. Communicative autonomy is severely limited (Ding & Shen, 2022). When schools ignore oral skills, learners increasingly look for alternative language applications online (Liu, 2025).

Social Media-Mediated Language Learning

While traditional classrooms often confine students to rigid grammatical instruction, Razali (2021) emphasizes that effective language acquisition relies on exposure to meaningful, authentic discourse. In this context, digital platforms may provide sustained exposure to comprehensible input (Krashen, 1985). Applications such as TikTok and YouTube provide learners with continuous access to authentic English and diverse accents. Pham, Van Nguyen & Cao (2023) demonstrate that by engaging with and imitating these multimedia inputs, learners acquire natural phrasing and intonation patterns that are frequently absent from standardized test preparation materials. Beyond receptive exposure, viewed through the lens of the Interaction Hypothesis, Karjagdi Çolak (2024) and Men (2023) assert that participating in digital interactions—such as leaving comments or recording short videos—provides the low-stakes practice opportunities that formal classrooms often lack. These environments also serve a clear psychological purpose. They easily lower the students' Affective Filter (Parent, 2023; He, 2024). Heavy exam pressure dominates the traditional classroom, but social media provides a clear alternative. This digital environment indicates a potential shift in the learning dynamic, as it effectively mitigates speaking anxiety and facilitates more spontaneous oral production (Alsharari, & Mohammed, 2026).

Dimensions of Speaking Development

In large classes, institutional assessment demands that prioritize standardized, easily graded formats may result in instructional time being devoted primarily to written exercises and structural drills, with limited opportunities for oral production. Students master structural rules but cannot perform basic communicative tasks (Ding & Shen, 2022). This strict focus on accuracy causes fear. It creates a 'freeze effect' and raises the Affective Filter, which significantly impedes language teaching (He, 2024).

To circumvent this communicative anxiety, learners increasingly gravitate towards social media platforms. Textbooks often feature stiff dialogue, so students use social apps to hear natural daily speech and slang (Razali, 2021; Mudhofar, 2023). Long (1996) argues that actual communication is necessary for speaking development. Since traditional, Exam-focused classrooms rarely provide this, digital platforms act as a necessary backup. By leaving comments and making videos, students easily find the low-pressure interaction they need.

Personalized Learning in Digital Environments

Digital platforms may facilitate forms of personalized learning that are less feasible within rigid classroom structures. Instead of following a one-size-fits-all curriculum, students can select videos and materials that match their specific interests. Dewi et al. (2023) highlight this advantage, viewing these online spaces as a practical way for students to take control of their own learning. This is an important shift because, as Ding and Shen (2022) point out, students are much more likely to practice speaking when they are genuinely interested in the topic they are discussing. Conventional classroom structures often require uniform pacing and standardized materials. However, social media provides a different approach. It allows learners to replay a short video as many times as they need or follow content that keeps them engaged. Furthermore, Fan (2023) and Mu’aggil and Pido (2023) note that this kind of freedom naturally increases student motivation and effort. Beyond learner engagement, Udam and Mohammed (2023) found that social media use contributed positively to teachers’ effectiveness, further demonstrating its potential to support more flexible and effective teaching and learning practices. Similarly, Alsharari and Mohammed (2026) found that the learning benefits of social media vary across platforms, suggesting that effective digital learning depends not merely on access to technology but on how particular platforms align with learners’ preferences, perceptions, and socio-cultural contexts. Taken together, these findings suggest that the value of social media in personalized learning lies in its capacity to provide flexible, context-sensitive learning experiences that can enhance both learner engagement and instructional effectiveness.

Methods

This study employed a structured mixed-methods approach to examine how social media influences EFL learners’ speaking development and how it compensates for limitations in exam-oriented classroom instruction. The methodological design was aligned closely with the two research questions to ensure both analytical depth and contextual relevance.

Research Design

The study applied a concurrent mixed-methods approach, collecting quantitative and qualitative data simultaneously. Creswell & Clark (2017) describe mixed-methods research as an approach that integrates numerical patterns with personal experience. Speaking development involves both measurable change and individual perception. Survey data provided quantitative patterns of reported improvement, while interview data offered qualitative explanations for these patterns.

Research Question 1 is answered using the survey data on fluency, pronunciation, and confidence. To address Research Question 2, the interviews explore social media as an out-of-class support. By employing data triangulation, the integration of survey and interview data offers a holistic understanding of the research problem, ultimately enhancing the methodological validity and analytical depth of the study (Creswell & Clark, 2017).

Research Context and Participants

The study was conducted in Guangxi, China, a region heavily driven by the Gaokao exam system. Ninety students aged 13 to 18 took the survey. Most of these participants had studied English for more than four years. Following the survey, five students were purposively selected for interviews to represent diverse urban and suburban backgrounds and varied social media habits. To maintain research integrity and mitigate potential risks, written informed consent was obtained from all participants prior to data collection, with personal identifiers removed to guarantee absolute anonymity in alignment with institutional ethical guidelines.

Questionnaire Development

Scholars such as Aloraini and Cardoso (2020), Razali (2021) evaluate digital language learning through core dimensions, including fluency, pronunciation, confidence, and learner autonomy. They identify these elements as critical factors for successful language learning. The questionnaire was developed based on these four dimensions. It targeted the influence of social media on speaking skills in contexts characterized by exam-focused instruction.

For the question format, Fauziah & Sutrisno (2024) present a simple Likert scale as the standard tool. In their view, it reliably measures how students feel about their progress. A five-point Likert scale was employed for data collection. The questionnaire underwent expert review to ensure clarity and comprehensibility prior to distribution.

Interview Sampling and Procedure

Creswell & Clark (2017) argue that exploring survey data requires participants from diverse backgrounds. Based on this idea, five students were selected for the interviews to capture a broad range of experiences. The participants included a student from a top city school, one from the suburbs, and others who use English apps primarily for recreational purposes. For the interview format, Dewi et al. (2023) recommend a semi-structured style. They recommend a flexible conversation format to capture detailed personal experiences related to online learning. A semi-structured format was employed for the interviews. Each session lasted between 20 and 30 minutes. The discussions not only addressed changes in pronunciation and confidence but also examined students’ experiences using English-learning applications in comparison with traditional exam-focused classrooms. All sessions were recorded and fully transcribed for analysis.

Data Collection Procedure

Creswell & Clark (2017) argue that survey results should guide the follow-up interviews. The study followed this sequence in data collection, beginning with the questionnaire. It was distributed online, enabling the collection of 90 responses from students who frequently use mobile devices. Once the main survey trends emerged, five students were invited to the interviews. These 20-minute sessions were scheduled during lunch breaks and weekends to accommodate students’ preparation for the Gaokao. All participant information was kept completely private. This secure setup allowed students to feel safe discussing their actual experiences. They provided candid responses regarding their academic pressure, openly sharing how they use these platforms.

Descriptive statistics from the 90 survey responses were analyzed to track frequency patterns related to pronunciation and confidence, an analytical approach that aligns with Fauziah and Sutrisno (2024). Interview data were analyzed through manual thematic analysis following Braun and Clarke (2006). Repeated readings of the transcripts moved the focus from general answers to specific initial codes, which were then systematically grouped into overarching themes such as pronunciation awareness and speaking confidence. The integration of quantitative survey results with qualitative interview themes provided a comprehensive account of students’ speaking development and daily learning experiences

Results and Discussion

An initial review of the survey data shows a broadly positive pattern across all 90 students. Most participants linked their social media use to better-spoken English. These shifts are detailed in Table 1, with a strong focus on fluency and pronunciation awareness. The data also reveals a higher general interest in learning.

Table 1: Students’ Perceptions of Social Media on Oral Proficiency

Categories	Survey Items	%Agreed/Strongly Agreed
Usage & Preference	Use social media to practice speaking Prefer social media over textbook drills	72.22% 67.78%
Fluency & Skills	Social media helps improve pronunciation My speaking speed has improved	75.56% 69.00%
Affective Factors	Pause or hesitate less when speaking Feel more confident online than face-to-face Feel less nervous in class after online practice	62.22% 70.00% 65.56%

Social Media as an Informal Speaking Extension

Both the survey and interviews point to a noticeable shift in how students approach daily language practice. Specifically, the quantitative data reveals that a substantial 72.22% of the participants actively use social media for speaking practice. Furthermore, 67.78% stated they prefer these digital platforms over traditional textbook drills. These figures highlight a strong preference for informal digital ecosystems over rigid classroom routines.

The interview sessions offer deeper qualitative insights into this behavioral shift. For many students, digital platforms—such as short videos and interactive clips—frequently replaced structured textbook exercises because they offer more engaging and relevant practice. However, the qualitative data also shows that this transition is not absolute; some students expressed a continued preference for the familiarity of traditional textbooks when reviewing grammar. This mixed picture makes sense alongside recent literature; the students' ability to select their own content aligns with Wardani (2023) and Pham, Van Nguyen & Cao (2023), who suggest that digital platforms enhance learner autonomy. As Razali (2021) emphasizes the necessity of meaningful communication, social media clearly offers these specific learners the vital communicative practice required to balance out standard, exam-oriented classroom instruction (Wardani, 2023).

Repetition and Implicit Pronunciation Training

Both the survey and interviews highlight strong perceived improvements in specific speaking skills. Specifically, the quantitative data reveals that 75.56% of respondents believe social media helps improve their pronunciation. Crucially, this digital exposure also appears to enhance oral fluency, as 69.00% felt their speaking speed had increased, and 62.22% noted they pause or hesitate less during speech.

The interview transcripts offer deeper qualitative insights into how this implicit training occurs. For many learners, speaking to a phone screen feels significantly less intimidating than speaking to a teacher who constantly monitors for mistakes. Students described these applications as an alternative space, separate from the strict demands of school schedules, where they can naturally pick up slang and real-world accents that rarely appear in traditional classrooms. These personal accounts illustrate how a lowered "Affective Filter" helps students relax and initiate speech, an observation that aligns with recent studies by He (2024) and Parent (2023). Furthermore, as Mudhofar (2023) and Wardani (2023) argue, informal practice naturally cultivates better conversational skills, indicating that students actively engage in meaningful communicative practice beyond the confines of exam-focused preparation.

Low-Stakes Practice and Confidence Development

Both the survey and interviews highlight social media as a safe, low-pressure space for language practice. Specifically, the quantitative data reveals that 70.00% of the surveyed students feel more confident practicing behind a screen than in face-to-face settings. Crucially, this online comfort appears to transfer to formal learning environments, as 65.56% of participants reported feeling less nervous during actual classroom activities after engaging in digital practice. These figures indicate a measurable reduction in speaking-related anxiety.

The interview sessions offer deeper qualitative insights into how this shift occurs. For many participants, the digital screen acts as a safe buffer that mitigates the immediate fear of public correction. For instance, watching "imperfect" English videos on Douyin shifted Hui's focus from grammatical perfection to actual communication, while digital platforms helped Lin view the language as a practical skill rather than just another exam requirement. These personal accounts illustrate how a low-stakes environment helps students build a stronger "English-speaking self," an observation aligning with recent studies by He (2024) and Fauziah & Sutrisno (2024). As Parent (2023) suggests, the confidence nurtured in informal digital spaces temporarily removes formal grading pressure and equips students with the courage needed for real-world situations, affording them the freedom to prioritize authentic communication.

Personalized Learning Within Exam-Oriented Systems

Both the survey and interviews demonstrate that social media can facilitate personalized practice outside formal classroom boundaries. Specifically, the quantitative data shows that a majority of the surveyed students feel these platforms helped them start thinking directly in English, rather than translating every word from Chinese. However, the survey also revealed a noticeable group of students who remained neutral about these benefits. This statistical gap suggests that digital access alone does not automatically guarantee effective learning. The interview sessions offer deeper qualitative insights into this discrepancy, revealing the crucial role of instructional guidance. Xiang’s experience provides a compelling example: she primarily used Xiaohongshu (Redbook) for entertainment until her teacher explicitly shared specific pronunciation videos on the app. Once she knew what to look for, the ability to utilize the "replay" function allowed her to engage in highly effective self-paced learning compared to traditional classroom lectures. These practical experiences highlight a key dynamic of digital learning: social media requires some level of direction to be fully effective, aligning with findings by Ding and Shen (2022). Xiang’s story adds specific contextual depth to the observations of Fan (2023) and Men (2023), appearing to indicate that a small amount of targeted teacher guidance transforms social media into a powerful supplementary educational tool.

Pedagogical Implications

The findings indicate that social media holds considerable potential; however, its actual impact appears to vary depending on students’ regional contexts. Even where hardware is available in rural schools, a persistent usage gap remains because, as Liu (2025) and Li et al. (2018) suggest, students often have the devices but don't know how to use them for more than just games. Therefore, the study prioritizes context-sensitive and practical strategies that can be realistically implemented in schools.

Instructional Practice: Adapting Tasks to Local Realities

Findings indicate that task design should consider school technology, as Dewi et al. (2023) suggest. For classrooms lacking individual devices, the main projector serves as the central learning tool. Teachers can introduce "Choral Dubbing." Instead of asking one student to speak alone, the entire class repeats a short video clip collectively. This collective practice effectively lowers speaking anxiety. It allows students to test pronunciation and rhythm without the pressure of individual performance.

In schools with better technology access, the main challenge is moving students from passive watching to active production. A 'Creative Output' approach works well for these specific classrooms. Students record 'contrastive role-play' videos to practice speaking. They first act out a common English mistake, and then they show the correct phrase. This specific task brings Çolak’s (2024) theory into actual classroom practice. It forces learners to be creators, so they carefully check their own language before speaking.

Curriculum Reform: Aligning Exams with Digital Literacy

The curriculum should connect formal exams with actual online communication. Achieving this requires a shift from static text to multimedia resources. Linking pop culture to the Gaokao syllabus offers a simple, practical bridge. Students analyze movie clips to find specific test phrases, and this serves as an effective alternative to traditional rote learning paradigms. This simply shows that real-world English and high-test scores can go hand in hand.

For students in rural areas with limited resource access, the curriculum should provide structured guidance. Li et al. (2018) notes that explicit direction works better than independent searching for these students. Putting QR codes in standard textbooks easily solves this problem. These codes can link directly to high-quality educational accounts. This official inclusion guides students to use their phones productively. As Mu’aggil & Pido, (2023) suggest, such integration turns a potential digital distraction into a valuable tool. It helps students balance their daily online habits with formal schoolwork.

Policy Advocacy: Formalizing Informal Digital Learning

Policy makers should do more than just approve digital learning. Since rural students often lack personal devices, converting current computer rooms into 'English-Speaking Labs' is a highly practical step. This approach follows Li et al., (2018), who argue that institutional resources are essential for supporting under-resourced learners. By opening these labs every week, schools give students a safe space to practice, which offers a low-cost method to improve digital access. Departments of education should also develop a 'cloud library' filled with micro-learning clips. Since long lectures are easily ignored today, these official resources need an influencer-like style. For example, a school platform could post simple 2-minute grammar 'hacks.' This method delivers the exact same curriculum, so students actually watch and learn. As Bu, Pan & Wang (2024) note, instructional clips should balance entertainment with academic accuracy. As Liu (2025) warns, without an official "content bank," rural students face a continuous disadvantage compared to urban peers who possess greater resource access. Therefore, providing targeted digital content is likely to help narrow the educational gap.

The data highlights a clear shift in student confidence, but the context remains highly specific. As the findings come entirely from rural Guangxi, they reflect a unique environment with limited resources. Urban learners typically have stronger digital support at home, so their online learning habits might look completely different. Also, a one-semester study mainly reveals short-term motivation. This short window captures the initial excitement of using social media, but it cannot tell us if these speaking habits survive the actual pressure of the final Gaokao.

Self-developed Instrument

The study utilized a self-developed questionnaire tailored to secondary schools in Guangxi. This design aligns with Li, Jee & Sun. (2018), who emphasizes the need for research tools that fit the reality of rural classrooms. While an expert review helped refine the initial questions, time and logistical limits prevented a formal pilot study. Omitting this step left the survey without a preliminary reliability test. Creswell & Clark (2017) observe that smaller studies frequently face this exact compromise. Still, without a trial run, certain items might have remained confusing for the students. A proper pilot study would have identified and resolved these hidden ambiguities prior to the main data collection.

Convenience Sampling and Sample Size

Relying on a convenience sample of 90 students represents a notable limitation of this study. Although the group included both urban and rural learners, the small sample size cannot fully represent the broader student population across Guangxi. In addition, voluntary participation introduces an inherent self-selection bias. Students who opted to join were likely not typical learners; they tended to be more comfortable with digital platforms and exhibited higher intrinsic motivation to practice English.

Aloraini & Cardoso, (2020) and Men (2023) identify this inherent self-selection bias as a pervasive challenge in digital learning research. Due to this bias, the positive findings might appear more optimistic than the reality of the general student body. Ultimately, these participants reflect the most active learners rather than the average student.

Regional Focus

Focusing exclusively on Guangxi limits the study to a highly specific linguistic and economic setting. The students' Zhuang or Cantonese backgrounds often caused them to mix up /n/ and /l/ sounds. Beyond pronunciation, Liu (2025) highlights the economic limits of southwestern China. Classroom observations often revealed groups of four or five students sharing a single borrowed smartphone to practice speaking. These exact technology and pronunciation issues differ from wealthier cities like Shenzhen, where students typically use personal devices and face different linguistic hurdles. The findings primarily apply to other under-resourced areas were strong local dialects and restricted screen access influence language acquisition outcomes.

Reliance on Self-reported Data

Rather than conducting objective speaking tests, the study based its findings entirely on self-reported data. Fauziah & Sutrisno (2024) and Pham, Van Nguyen & Cao (2023) note that this is a common challenge in this type of research. As a result, reported gains in fluency and speed depend entirely on student perception. Students often feel they have improved significantly once they gain speaking confidence. But this increased confidence does not automatically guarantee better grammar (He, 2024). Many participants became more fluent in class, and they still made basic errors. Without baseline pre-tests, true language development remains unmeasured. The findings simply highlight a boost in confidence, not verified linguistic progress.

Conclusion

Social media clearly acts as a practical tool in Guangxi. It connects exam prep to real communication, and this low-pressure practice reduces student anxiety. As a result, the classroom “freeze effect” was significantly mitigated. A distinct usage gap also appeared. While urban students use these tools to create their own content, rural students mostly remain passive watchers. Since giving out devices is not enough, teachers need to show students how to use them productively.

Looking ahead, future research must move beyond self-reported feelings. Due to the limited timeframe of the current research, it remains unclear whether this confidence persists during actual examinations. Finally, comparing developing regions like Guangxi to wealthier urban centers like Shanghai would provide valuable insights into how family income and home environments shape digital learning. In the end, the main goal goes beyond pushing technology into classrooms. It is about making sure geography no longer limits speaking opportunities.

Declarations

Ethics Approval and Consent to Participate: All participants provided informed consent before data collection. Their participation was entirely voluntary, and their responses were kept confidential throughout the study.

Conflicts of Interest: Not Applicable.

Acknowledgement: Heartfelt gratitude is extended to all participants for their meaningful contributions to this research.

References

Aloraini, N., & Cardoso, W. (2022). Social media in language learning: A mixed-methods investigation of students’ perceptions. Computer Assisted Language Learning, 35(8), 1707-1730. https://doi.org/10.1080/09588221.2020.1830804

Alsharari, T. F., & Mohammed, L. A. (2026). Exploring the role of socio-cultural factors and perceived usefulness in social media-assisted EFL learning: Evidence from Saudi Arabia. Scientific Culture, 12(2.1), 7993–8007. https://doi.org/10.5281/zenodo.19714121

Bu, Z., Pan, S., & Wang, Z. (2024). Enhancing Vocabulary Learning: Multimodal Engagement in English Teaching Social Media Video. Arts, Culture and Language, 1(6).

Creswell, J. W., & Clark, V. L. P. (2017). Designing and conducting mixed methods research. Sage publications.

Dewi, B. C., Agustiana, S., Riska, R., & Adiantika, H. N. (2023). THE UTILIZATION OF SOCIAL MEDIA IN EFL HIGHER EDUCATION CLASSROOM CONTEXT. Indonesian EFL Journal, 9(1), 75-82. https://doi.org/10.25134/ieflj.v9i1.7157

Ding, Y., & Shen, H. (2022). Delving into learner autonomy in an EFL MOOC in China: A case study. Computer Assisted Language Learning, 35(3), 247-269. https://doi.org/10.1080/09588221.2019.1681464

Fan, C. (2023). English learning motivation with TAM: Undergraduates’ behavioral intention to use Chinese indigenous social media platforms for English learning. Cogent Social Sciences, 9(2), 2260566. https://doi.org/10.1080/23311886.2023.2260566

Fauziah, A., & Sutrisno, D. (2024). High school students' perceptions of using social media in learning English: A quantitative study. Teaching English as a Foreign Language Journal, 3(1), 52-67. https://doi.org/10.12928/tefl.v3i1.974

He, H. (2024). Analyzing the relationship between social media usage and foreign language learning anxiety among Chinese college students: A quantitative study. https://gupea.ub.gu.se/bitstream/handle/2077/82053/Studentessay_VT24_PDA184_HH.pdf

Karjagdi Çolak, M. (2024). Pathways to motivation and learning in English language education: a case of two youths engaging in multiliteracies. English Teaching: Practice & Critique, 23(4), 442-462. https://doi.org/10.1108/ETPC-11-2023-0148

Krashen, S. D. (1985). The input hypothesis: Issues and implications. (No Title). https://ejournal.uinsgd.ac.id/index.php/educater/article/view/2053

Li, G., Jee, Y., & Sun, Z. (2018). Technology as an educational equalizer for EFL learning in rural China? Evidence from the impact of technology-assisted practices on teacher-student interaction in primary classrooms. Language and Literacy, 20(3), 159-184. https://doi.org/10.20360/langandlit29415

Liu, G. L. (2025). Digital empowerment for rural migrant students in China: Identity, investment, and digital literacies beyond the classroom. ReCALL, 37(2), 215-231. https://doi.org/10.1017/S0958344024000302

Long, M. (1996). The role of the linguistic environment in second language acquisition. Handbook of second language acquisition.

Men, Y. (2023). The multimodal English learning strategies, motivation, and engagement among college students in China. Journal of Education and Educational Research, 4(2), 116–120. https://doi.org/10.54097/jeer.v4i2.10834

Mudhofar, A. F. (2023) The utilization of Tiktok for oral English competence of EFL students. https://eprints.walisongo.ac.id/id/eprint/23843/

Mu’aggil, I. M. R., & Pido, N. W. T. (2023). Social media as learning tools: a bibliometric study of educational applications. Journal of English Teaching and Linguistic Issues (JETLI), 2(2), 65-74. https://doi.org/10.58194/jetli.v2i2.1835

Parent, N. (2023). Basic need satisfaction through social media engagement: A developmental framework for understanding adolescent social media use. Human Development, 67(1), 1-17. https://doi.org/10.1159/000529449

Pham, M. T., Van Nguyen, T. T., & Cao, T. X. T. (2023). Vietnamese students' perceptions of utilizing social media to enhance English vocabulary: A case study at Van Lang University. International Journal of TESOL & Education, 3(1), 79-111. https://doi.org/10.54855/ijte.23316

Razali, A. B. (2021). An overview of the utilization of TikTok to improve oral English communication competence among EFL undergraduate students. Universal Journal of Educational Research. https://sl1nk.com/s7roc7s

Udam, M. U., & Mohammed, L. A. (2023). Impact of Social Media on Science Teacher’s Effectiveness in Secondary Schools in Cross River State. International Journal of Emerging Issues in Social Science, Arts and Humanities (IJEISSAH), 2(1), 89-100. https://doi.org/10.60072/ijeissah.2023.v2i01.010

Wardani, K. D. K. A. (2023). Language Learning Through Social Media: Insight from Bibliometric and Content Analysis. In ICHSS 2022: Proceedings of the 2nd International Conference of Humanities and Social Science, ICHSS 2022, 17 December 2022, Surakarta, Central Java, Indonesia (p. 11). European Alliance for Innovation. https://doi.org/10.4108/eai.17-12-2022.2332889