A Qualitative Thematic Analysis of ChatGPT-4 Integration in Extensive Reading to Enhance Critical Thinking in EFL Higher Education

Critical Thinking in a GenAI World: Evaluating the Efficacy of ChatGPT-Assisted Essay Writing in ESL Classrooms

The brisk integration of Generative Artificial Intelligence (GenAI), particularly Large Language Models (LLMs) like ChatGPT, has fundamentally disrupted educational paradigms, specifically second language acquisition (SLA) and English as a Second Language (ESL) writing instruction. While GenAI tools offer unparalleled scaffolding for grammar, vocabulary generation, and structural organisation, their impact on higher-order cognitive skills remains highly contested. This study evaluates the empirical efficacy of ChatGPT-assisted essay writing in university-level ESL classrooms, focusing specifically on its effects on students’ critical thinking (CT) skills. Using a mixed-methods convergent parallel design over a 15-week academic semester, this study examined $N = 124$ undergraduate ESL students split into an experimental group (utilising ChatGPT as a structured co-writing and dialogue partner) and a control group (utilising conventional process-writing methodologies). Quantitative data were gathered using pre- and post-test assessments scored via a validated Critical Thinking Rubric (evaluating argument analysis, evidence evaluation, alternative perspectives, and bias recognition) along with automated linguistic analysis of student essays. Qualitative data were obtained via semi-structured focus group interviews and student reflective journals. The quantitative findings indicate that while the experimental group demonstrated significant improvements in surface-level linguistic accuracy, lexical density, and structural cohesion, their independent scores in evidence evaluation and independent thesis formulation were statistically lower than those of the control group when ChatGPT assistance was removed in unassisted proctored post-tests ($p < .05$). The qualitative thematic analysis revealed a phenomenon termed "cognitive outsourcing," where students routinely accepted AI-generated assertions without epistemological verification. However, a sub-cohort trained in aggressive prompt engineering and dialectical inquiry demonstrated elevated critical engagement. This paper proposes a “Dialectical AI-Scaffolding Framework” designed to shift ChatGPT from a product-generation tool to an intellectual sparring partner, ensuring that GenAI advances rather than compromises critical thinking in ESL pedagogy.

ChatGPT-Mediated Materials and Assessment Design in an Omani EFL Foundation Programme: An Intelligent-TPACK Case Study

Generative artificial intelligence can accelerate English-as-a-foreign-language (EFL) materials and assessment design, yet its pedagogical value depends on teachers’ capacity to evaluate linguistic quality, contextual appropriateness, and ethical risk. This interpretive case study examined how five General Foundation Programme EFL teachers at Sohar University, Oman, used ChatGPT to prepare instructional materials and design assessments. Semi-structured interview data were analysed deductively in NVivo through the Intelligent-TPACK framework and Braun and Clarke’s six-phase thematic procedure. The teachers used ChatGPT to develop speaking and writing tasks, reading passages, vocabulary and grammar exercises, lesson ideas, and formative assessments. Effective use was described as an iterative process involving explicit prompts, proficiency-level specifications, learning outcomes, templates or tables of specifications, output checking, and prompt refinement. Participants valued efficiency, ideation, and adaptability, but reported concerns about inaccurate or unreliable content, weak coherence and authenticity, cultural inappropriateness, data privacy, academic integrity, overreliance, and the possible erosion of teacher and learner creativity. The study shows that productive teacher–AI interaction is not one-step automation but a cycle of pedagogical specification, linguistic validation, contextual adaptation, and ethical judgement. It recommends targeted professional development and institution-level guidance for responsible generative-AI use in language teaching and assessment.

Students' experiences of using ChatGPT for English language learning: a qualitative study in a Malaysian higher education institution

A qualitative thematic analysis was conducted with 25 students from a Malaysian institution to gain a deeper understanding of their motivation for using ChatGPT in English language learning. The study employed Self-Determination Theory (SDT) and the results demonstrated that ChatGPT has the potential to contribute positively to aspects related to English language competence, social connections, and independent learning processes. Therefore, ChatGPT emerges as a helpful resource for enhancing grammar, writing, and conversational tasks, allowing educators to prioritize higher order thinking abilities and intricate areas of language training. The results suggest that ChatGPT supports learners' autonomy, relatedness, and competence, which are key components of Self-Determination Theory (SDT). Nevertheless, the study also highlighted certain obstacles, such as ChatGPT's tendency to produce erroneous or nonsensical information and its restricted capacity to ensure the accuracy of content in language learning situations. These findings emphasize the need for focused efforts to tackle these difficulties. Educators must acknowledge the limitations of the tool and complement ChatGPT interactions with human interactions to provide a comprehensive and efficient language learning experience.

Adapting Literature Circles for Critical Thinking in Indonesian EFL Classrooms: A Qualitative Case Study Aligned with SDG 4

Objective: To investigate how Literature Circles can be pedagogically adapted to the Indonesian English as a Foreign Language (EFL) context to foster students' critical thinking while accommodating curriculum requirements, students' language proficiency, and classroom learning needs in support of Sustainable Development Goal (SDG) 4. Method: A qualitative case study was conducted at an Indonesian senior high school involving one English teacher and twelve eleventh-grade EFL students. Data were collected through classroom observations, semi-structured interviews, and document analysis, including lesson plans, students' worksheets, dialogue journals, and learning products. The data were analyzed using coding, categorization, thematic analysis, and triangulation. Results: The findings revealed that Literature Circles were successfully adapted through modifications to instructional procedures, collaborative learning stages, reading materials, and role assignments, including the addition of Text Analyzer and Grammar Pointer roles to align with the Indonesian genre-based curriculum. These adaptations promoted six dimensions of critical thinking interpretation, analysis, inference, evaluation, explanation, and self-regulation while enhancing students' affective, cognitive, and behavioral engagement in collaborative learning. Novelty: The study proposes a contextualized adaptation model of Literature Circles for Indonesian secondary EFL classrooms, demonstrating how collaborative reading can be integrated with curriculum demands to foster critical thinking, student engagement, and the achievement of Sustainable Development Goal 4 (SDG 4) through student-centered learning.

Language and Communication Students' Perceived Mastery in AI Chatbot Prompt Engineering: A Study on Vibe Coding in Educational Mobile Application Development

This study examines language and communication students’ self-perceived mastery of AI chatbot prompt engineering, with particular attention to “vibe coding” for educational mobile application development. In this study, vibe coding refers to the deliberate use of prompts to shape tone, style, audience orientation, and user-facing educational content. Using a pattern-based framework centred on roles, constraints, and examples, the study investigates how students combine structure and exploratory prompting. A quantitative cross-sectional survey was administered online during March–July 2025 (Semester 2). Complete responses from 120 undergraduates were analysed from approximately 126 invited students (analytic response rate = 95.2%). The questionnaire demonstrated excellent overall internal consistency (Cronbach’s ? = .965). Results indicate that students usually begin tasks with structured prompts but later move towards mixed or unstructured prompting styles, suggesting a control-then-explore sequence. Longer exposure to AI chatbots and more extensive prompt-engineering training were associated with higher self-perceived competency and output efficiency, whereas weekly usage frequency did not show statistically significant differences. Educational background was associated with prompting style and usage, but not with overall perceived competency or output efficiency, and gender differences were negligible. The findings suggest that scaffolded instruction in prompt engineering can support more confident and consistent student use of AI chatbots. Future research should triangulate self-reports with behavioural logs, archived prompts, and performance-based outputs.

Adapting Literature Circles to Foster Critical Thinking Skills in Indonesian EFL Classrooms: A Qualitative Case Study toward SDG 4