MINISTRY OF EDUCATION AND TRAINING HO CHI MINH CITY OPEN UNIVERSITY ********* PHAM MINH TOAN ELT TEACHERS’ ACCEPTANCE AND APPLICATION OF CHATGPT AS A WRITING ASSISTANT IN GENERAL ENG
INTRODUCTION
Background of the Study
Writing skills are essential for language learning and effective communication, blending creativity, linguistic knowledge, and critical thinking In EFL contexts, teachers often encounter difficulties helping students produce accurate and meaningful texts due to limitations in vocabulary, grammar, and contextual understanding These challenges are intensified by the cognitive demands of organizing ideas and creating coherent writing In Vietnam, significant differences between English and Vietnamese linguistic structures necessitate tailored and culturally relevant teaching approaches.
AI tools like ChatGPT have revolutionized education by providing innovative solutions that enhance writing instruction for both teachers and students These tools offer instant feedback, vocabulary suggestions, and grammar corrections, significantly improving writing proficiency and engagement (OpenAI, 2023; Sumakul et al., 2022; Marzuki et al., 2023) Despite their potential, research has primarily focused on student outcomes, with limited exploration of teachers' perceptions and their integration of such technologies Challenges like biases, contextual misunderstandings, and the necessity for thorough review of AI outputs may hinder teachers' adoption of ChatGPT (Tatar et al., 2024) This underscores the importance of addressing these concerns to facilitate effective use of AI in educational practices.
2 professional development and clear strategies to support effective AI integration in teaching practices
ChatGPT offers considerable advantages for educators by reducing workloads, fostering creativity, generating writing prompts, and delivering personalized feedback, thus serving as a valuable asset for writing instruction (Nguyen, 2023) However, there is limited understanding of English Language Teaching (ELT) teachers' perceptions and utilization of ChatGPT in Vietnam, highlighting the need to explore how these educators view, prepare for, and integrate this AI tool into their teaching methodologies.
Acknowledging the importance of writing skills and the growing influence of AI in education, the author aims to investigate English Language Teaching (ELT) teachers' acceptance and use of ChatGPT as a writing assistant in English courses, while also addressing the challenges educators encounter in integrating AI tools and the scarcity of research on teachers' perspectives.
Statement of the problem
The integration of AI technologies like ChatGPT has revolutionized education, particularly in language teaching, through its advanced natural language processing capabilities By offering personalized learning experiences, ChatGPT provides real-time feedback, corrects grammatical errors, and generates a variety of writing prompts, effectively enhancing both teaching and learning in writing instruction.
While ChatGPT has the potential to improve writing instruction, it poses challenges for English Language Teaching (ELT) teachers regarding readiness and classroom integration The AI's feedback often lacks the customization needed to meet individual student needs, leading teachers to invest additional time in refining these suggestions, which can diminish the tool's effectiveness (Yoon et al., 2023) Furthermore, an over-reliance on ChatGPT may reduce meaningful teacher-student interactions, as students who depend heavily on AI for feedback may obstruct teachers' ability to offer personalized support, ultimately weakening the teacher-student relationship (Baron, 2023; Whang, 2024; Widianingtyas et al., 2023).
Additionally, limited professional training and technical support may leave teachers feeling unprepared to navigate these challenges effectively and confidently (Tatar et al., 2024)
In my institution, the use of ChatGPT as a writing assistant in general English courses presents both advantages and challenges While it aids students in developing fundamental writing skills through clear structures, straightforward ideas, and correct grammar, it also provides rapid feedback that enables students to independently identify and correct errors (Guo & Wang, 2023) However, an overreliance on AI-generated feedback may hinder the development of critical thinking skills, impacting teachers' efforts to promote independent thought Furthermore, some educators express reservations about fully integrating ChatGPT, as its suggestions may not always align with a curriculum focused on personalized instruction and student-centered learning Therefore, teachers must thoughtfully evaluate how ChatGPT aligns with their instructional goals, influencing their acceptance and integration of this tool into their teaching practices.
Current research on the application of AI, especially ChatGPT, in writing education is still developing Globally, investigations have primarily concentrated on student outcomes, while there has been a lack of in-depth analysis regarding teachers' views on the acceptance and utilization of these technologies.
Research on the integration of AI tools, particularly ChatGPT, into English writing instruction in Vietnam is limited, especially concerning the acceptance and application among English Language Teaching (ELT) teachers (Nguyen, 2023) This study aims to fill this critical gap by investigating how ELT teachers in Vietnam accept and utilize ChatGPT, highlighting the unique context that has been overlooked in international research By situating this study within the global discourse on AI in education, it contributes to a deeper understanding of local practices while enriching the international literature on the subject.
Teachers encounter both challenges and opportunities when incorporating AI tools into writing instruction Understanding these insights is crucial for enhancing future research and practical applications of AI in English teaching worldwide.
In light of the identified issues and gaps in existing research, this study aims to explore "ELT Teachers’ Acceptance and Application of ChatGPT as a Writing Assistant in General English Courses."
Research Aims and Questions
This study aims to explore the ELT teachers’ acceptance and application of ChatGPT as a writing assistant in general English courses To be more specific, this study aims:
1 To explore the ELT teachers’ acceptance of ChatGPT as a writing assistant in general English courses
2 To explore the ELT teachers’ application of ChatGPT as a writing assistant in general English courses
This research attempts to seek the answer to the following research questions:
1 How do teachers accept ChatGPT as a writing assistant in general English courses?
2 How do teachers apply ChatGPT as a writing assistant in general English courses?
Scope of the Study
The study investigated the acceptance and application of ChatGPT as a writing assistant by ELT teachers in general English courses at College A in Ho Chi Minh City, Vietnam, involving six teachers from April to September 2024 Utilizing a single-method qualitative approach, the research employed semi-structured interviews to gather insights The theoretical framework was grounded in the Technology Acceptance Model (TAM) proposed by Davis (1989), focusing on key themes such as ease of use, context, and methodology.
5 usefulness, and acceptance, and showed how teachers accept and apply ChatGPT to support and enhance student writing.
Significance of the Study
This study investigates the acceptance and utilization of ChatGPT by English Language Teaching (ELT) instructors as a writing assistant in general English courses, highlighting its potential to support teachers in fostering students' writing skill development.
This thesis aims to expand the existing research on teachers' acceptance and use of ChatGPT as a writing assistant in English Language Teaching (ELT) While much of the international focus has been on student outcomes related to AI tools like ChatGPT, this study emphasizes the perspectives and experiences of teachers, an essential yet underrepresented area By examining the acceptance and application of ChatGPT among ELT teachers in Vietnam, the research addresses a significant gap in global literature, offering valuable insights from a cultural and educational context that has been largely overlooked.
This study aims to explore how teachers utilize ChatGPT in general English courses to improve their teaching methods By focusing on practical strategies and challenges, the research seeks to enhance the understanding of AI tool integration in education globally The findings will serve as a valuable resource for policymakers and educators looking to adapt AI tools to various educational contexts.
This research enhances the author's understanding of educational technology, particularly ChatGPT, and its influence on language teaching Engaging with educators reveals critical insights into the challenges of integrating ChatGPT, thereby improving both research skills and practical knowledge By focusing on real-world applications, the study ensures that its contributions are practical and theoretical, laying a solid foundation for future research and promoting the development of effective AI-driven educational tools.
This article explores six effective teaching practices utilized worldwide, highlighting the author's commitment to leveraging technology to enhance language education By integrating innovative approaches, these practices significantly enrich academic and instructional methods, particularly in writing instruction.
Organization of the Thesis
This thesis comprises five chapters that explore various dimensions of English Language Teaching (ELT) teachers' acceptance and use of ChatGPT as a writing assistant Chapter One sets the stage by detailing the research context, objectives, and key questions, while emphasizing the study's importance and offering a comprehensive overview of the thesis structure.
Chapter Two examines theoretical frameworks and prior research regarding the use of ChatGPT in education, with a specific focus on its role in teaching writing Additionally, it assesses teachers' acceptance of this tool and highlights existing research gaps that the current study aims to address.
Chapter Three explains the research design, sampling methods, and data collection procedures, which emphasizes the qualitative approach It also describes the process of analyzing the collected data
Chapter Four presents the key findings derived from thematic analysis The discussion connects the findings with the reviewed literature to provide deeper insights
Chapter Five discusses their broader implications for teaching and learning It also acknowledges the limitations of the study and suggests potential directions for future research.
Chapter Summary
The first chapter introduces the study on English Language Teaching (ELT) teachers' acceptance and utilization of ChatGPT in writing instruction, highlighting both the advantages and challenges of its integration It outlines the study's objectives, research questions, and significance while also offering a roadmap for the thesis structure.
LITERATURE REVIEW
ChatGPT as an AI Tool
This section explores ChatGPT as an AI tool, starting with a brief history of artificial intelligence, then examining its use in language learning via chatbots, and finally providing an overview of ChatGPT's key features and development.
2.1.1 Evolution and Impact of AI and Chatbots in Language Learning
The journey of artificial intelligence (AI) began with the goal of replicating human intelligence, starting from basic algorithms and progressing to sophisticated machine learning systems The formal inception of the field occurred in 1956 at the Dartmouth Conference, where the term "Artificial Intelligence" was coined Significant advancements in computing power and data accessibility from the 1980s through the 2000s spurred rapid developments, enabling machines to tackle more complex challenges Today, AI has made a substantial impact across various sectors, particularly in education, where tools like ChatGPT have transformed language learning by facilitating more interactive and human-like conversations.
AI has paved the way for many future innovations such as AI chatbots, which play significant roles across various fields Chatbots are software designed to
8 simulate human conversation, which allows interaction with digital devices via text or voice (Niòen, 2021) Simple chatbots follow set rules, while advanced ones use
AI, NLP, and Machine Learning (ML) for more natural interactions (Suta et al., 2020)
AI-powered chatbots are revolutionizing language learning by delivering personalized and interactive experiences that adapt to individual learners' needs, offering instant feedback to enhance skill development (Nguyen, 2022; Chen et al., 2021) Tools like ChatGPT have proven particularly effective in promoting conversation practice and improving language proficiency.
ChatGPT, created by OpenAI, is a sophisticated AI chatbot recognized for its natural language processing (NLP) abilities and engaging interactions It is characterized as a large language model (LLM) utilizing a transformer-style neural network architecture and pre-trained on extensive text data to predict subsequent tokens in a document (Kaarre et al., 2023) OpenAI (2023) highlights that ChatGPT is designed to generate human-like text and facilitate natural, intelligent conversations Additionally, Kung et al (2023) note that ChatGPT is a recent general LLM developed by OpenAI, which predicts the probability of word sequences based on prior context, powered by the GPT-3.5 model, trained on a vast corpus of internet text through reinforcement and supervised learning techniques (p 126).
This research explores the acceptance and application of ChatGPT among English Language Teaching (ELT) teachers, emphasizing its role as a versatile educational tool Its significance in ELT is underscored by its language generation capabilities, which are crucial for enhancing student writing skills Consequently, the definition provided by Kung et al (2023), which characterizes ChatGPT as a "general" resource, is deemed the most appropriate for this context.
LLM, [ ], trained on a large corpus of text data, [ ], via reinforcement and supervised learning methods” (p 126), aligning with educational needs
ChatGPT is an effective tool for teaching and learning English, thanks to its natural language generation that produces coherent and contextually accurate responses, making it ideal for essay writing and grammar correction The tool's interactivity provides real-time feedback, enabling personalized conversations that help learners refine their skills Additionally, its customizability allows users to tailor responses for specific needs, such as writing prompts or practice tests While ChatGPT's large-scale training enhances its ability to process diverse inputs, it may occasionally produce inaccuracies due to its reliance on pre-existing data Overall, these features demonstrate ChatGPT's potential to improve writing skills and promote creative learning.
ChatGPT as a Writing Assistant Tool
2.2.1 Theoretical Foundations of ChatGPT as a Writing Assistant
This section delves into the theoretical underpinnings of ChatGPT as a writing assistant, emphasizing its function in scaffolding writing through the lens of social constructivism Scaffolding, a concept first introduced by Wood et al., plays a crucial role in enhancing the writing process.
Scaffolding, a crucial technique in English Language Teaching (ELT), facilitates learners' skill development by offering temporary support, in line with Vygotsky’s Zone of Proximal Development (ZPD) theory Developed in the 1930s and published in "Mind in Society" (Vygotsky, 1978), this theory, grounded in social constructivism, highlights the significance of guided assistance and social interaction in enabling learners to achieve tasks beyond their independent capabilities As learners gain proficiency, this support is gradually diminished (Wood, Bruner, & Ross, 1976; Rogoff, 1991), effectively clarifying learning challenges and enhancing the educational experience.
ChatGPT enhances higher-order thinking and positively influences classroom dynamics by reducing anxiety, promoting communication, and fostering collaboration It aligns with proven scaffolding practices by offering timely feedback, generating model texts, facilitating reflective discussions, and providing language support These connections are crucial as they illustrate how ChatGPT can strengthen teachers' abilities to effectively support student writing The next section will analyze the theoretical foundations for utilizing ChatGPT as a writing assistant.
When it comes to the theories relevant to using ChatGPT as a writing assistant, there are several scaffolding strategies that align with their role in supporting writing development
ChatGPT supports teachers by implementing cognitive scaffolding theory, emphasizing the importance of timely feedback in improving the writing skills of EFL learners Research by Cheng and Zhang (2022) indicates that effective feedback enables educators to identify and address students' writing weaknesses, ultimately helping them achieve their desired proficiency levels.
Effective feedback is crucial for guiding instructional adjustments to improve teaching outcomes The instant and personalized feedback provided by ChatGPT aligns with cognitive scaffolding theory, which highlights the importance of timely support during learning By offering real-time corrections and suggestions, ChatGPT assists teachers in bridging the gap between learners' current skills and their learning objectives This ongoing feedback not only enhances students' grammatical accuracy but also promotes learner engagement by creating a responsive learning environment.
ChatGPT is an essential resource for educators, facilitating the implementation of content scaffolding by providing model texts tailored to diverse student proficiency levels Incorporating model essays in English Language Teaching (ELT) significantly enhances multiple facets of writing, such as accuracy, complexity, vocabulary, grammar, discourse, and overall content quality (Dizaj & Karbalaei, 2014; Saeidi & Sahebkheir, 2011; Wu et al.).
Model essays serve as valuable feedback tools for teachers, providing well-structured examples that enhance students' writing performance (Wu et al., 2023) By utilizing these examples, educators can effectively guide learners to recognize essential language features, including vocabulary and organization, allowing students to integrate these elements into their own writing (Saeidi & Sahebkheir, 2023).
ChatGPT aids teachers in providing meta-cognitive scaffolding, allowing students to discuss the creation and organization of their writing By facilitating discussions or debates on English writing assignments, teachers can significantly improve students' writing skills Kassem (2021) emphasizes that engaging students in fluent speaking activities enhances their perceived competence in interactions This highlights the importance of writing skills in shaping student interactions and responses to arguments Furthermore, Majidi et al (2021) show that integrating debates into EFL instruction positively impacts students' oral and written communication abilities.
ChatGPT is a valuable tool for teachers, offering linguistic scaffolding that enhances English Language Teaching (ELT) learners' writing skills and addresses cognitive and cultural challenges (Lincoln & Idris, 2015) It provides prompt linguistic feedback in line with social-constructivist principles, promoting learning through interaction (Punar ệzỗelik & Yangın Ekşi, 2024) This support allows educators to tailor instruction in grammar, sentence structure, and vocabulary to meet learners' diverse linguistic backgrounds, facilitating idea connection and overcoming language barriers (Torff & Murphy, 2020) However, the effectiveness of ChatGPT depends on teachers' acceptance and integration of the tool, which is crucial for maximizing its potential in language education The next section will explore several key functions of ChatGPT as a writing assistant.
2.2.2 Key Functions of ChatGPT as a Writing Assistant
In language teaching, ChatGPT serves as a valuable writing assistant, aiding learners in idea generation, proofreading accuracy, and fostering collaboration This AI tool offers customizable feedback to address diverse learning needs and provides advanced functions like producing consistent content, editing, and instant feedback, making it an essential asset for enhancing writing skills and improving instructional effectiveness.
ChatGPT serves as a valuable writing assistant by generating innovative ideas and suggesting topics and themes that educators can present to students, enhancing their creative thinking and broadening their writing horizons (Kasneci et al., 2023; Taecharungroj).
In 2023, tools like ChatGPT provide valuable prompts that help teachers tackle challenges in idea generation, allowing them to concentrate on more complex writing tasks By offering diverse suggestions and perspectives, ChatGPT supports educators in fostering creativity and structure in their students' writing However, it is crucial to incorporate human refinement to maintain the originality and depth of AI-generated content.
ChatGPT serves as a valuable writing assistant for teachers by enhancing the accuracy and consistency of their content With its vast information access, it effectively identifies and corrects errors, ensuring high-quality instructional materials and feedback (Stacey, 2022) This tool aids educators in maintaining uniformity in format, language, and style, allowing them to provide clear and well-structured examples that significantly improve students' writing skills (Cacicio).
ChatGPT serves as an efficient tool for proofreading and editing, aiding teachers in helping students improve their grammar, sentence structure, and spelling, ultimately leading to higher-quality writing with fewer errors It can also pinpoint minor clarity and coherence issues, enabling students to refine their arguments and enhance overall readability Furthermore, its ability to analyze context and provide targeted suggestions allows teachers to deliver more constructive feedback in a shorter amount of time.
ChatGPT enhances collaboration between teachers and students by providing instant feedback on writing, promoting active participation, and encouraging self-directed learning This tool keeps learners focused on their writing tasks and allows them to identify and correct mistakes in real-time, leading to a better understanding of writing conventions The interactive support offered by ChatGPT not only improves writing accuracy but also boosts learners' confidence in producing high-quality work independently.
Teacher Acceptance of ChatGPT
Before exploring teachers' acceptance of ChatGPT, it's crucial to define "acceptance" as "users' willingness to use a system" (Dillon & Morris, 1996) This concept encompasses a prospective judgment regarding future measures, indicating that the target group has not yet experienced the new technology (Schade & Schlag, 2003) Adell (2010) further clarifies acceptance as the degree to which individuals intend to use and incorporate a system into their practices Similarly, Proctor et al (2011) describe acceptance as the perception among stakeholders that a given innovation is agreeable and satisfactory Understanding acceptance lays the groundwork for analyzing teachers' readiness to adopt new technologies in educational settings.
Some authors highlight the significance of timing in the technology acceptance process Martin et al (2015) define "technology acceptance" as an individual's perception of a system prior to its use Additionally, Overdijk and Van Diggelen (2006) describe it as a social construction process where user actions and thoughts are influenced by technology, while the technology's meaning and effects evolve through user interactions.
(1989) also explained “technology acceptance” as “the degree to which a person believes that using a particular system would enhance their job performance” (p 320)
He further described the system as “the degree to which a person believes that using a particular system would be free of effort” (p 320)
Technology acceptance, as defined by various scholars, generally pertains to the adoption and integration of new technologies For this study, which focuses on teachers' perceptions, Davis's (1989) concept is particularly relevant and serves as the foundational framework for understanding "teacher acceptance." In essence, "teacher acceptance" reflects the willingness of educators to embrace and incorporate new technologies into their teaching practices, shaped by their experiences and beliefs.
Teachers' acceptance of incorporating ChatGPT into writing instruction hinges on their perceptions of its ease of use and perceived usefulness This acceptance goes beyond mere openness to technology; it involves acknowledging its potential to improve instructional effectiveness in teaching writing The subsequent section explores the key factors that influence teachers' acceptance of ChatGPT as a valuable writing assistant.
2.3.2 Teacher Acceptance of ChatGPT in Language Teaching
Teachers' acceptance of ChatGPT for writing instruction is significantly influenced by its perceived ease of use and perceived usefulness These two factors play a crucial role in determining how readily teachers embrace ChatGPT in their teaching practices The tool's simplicity and accessibility enhance its perceived ease of use, enabling educators to incorporate it into their lessons more efficiently.
16 effectively (Strzelecki et al., 2024) Furthermore, its perceived usefulness, such as reducing workload and providing instant feedback, enhances the motivation to accept
To begin with, perceived ease of use affects teachers’ acceptance of ChatGPT because of its easy access and instant feedback According to Hasanein and Sobaih
In 2023, ChatGPT gained popularity in education due to its accessibility, requiring only an internet connection and no special technical skills Tarchi et al (2024) emphasize its user-friendly design, which facilitates generating writing prompts, checking grammar, and receiving feedback without extensive AI knowledge The immediate feedback on grammar, style, and structure provided by ChatGPT significantly enhances teacher acceptance of writing instruction, showcasing the platform's practical benefits for improving writing quality and efficiency (Kuziev, 2024) Additionally, this feedback not only aids the revision process but also encourages the exploration of diverse writing techniques (Solovey, 2024).
Motivation significantly influences teachers' acceptance of technology, particularly in enhancing perceived usefulness (Nguyễn, 2023) ChatGPT can notably boost teacher motivation in writing instruction by offering personalized feedback, enabling educators to concentrate on more engaging lesson elements (Baskara, 2023) Furthermore, Ali et al (2023) highlight that ChatGPT's capacity to generate writing prompts and model essays keeps students engaged while alleviating the pressure on teachers to develop new materials continually Additionally, real-time feedback from ChatGPT fosters increased student participation and allows for more personalized instruction, ultimately enhancing teachers' motivation over time (Kuziev, 2024).
Teacher acceptance of ChatGPT is shaped by various factors, including professional training programs, specific contextual needs, and the perceived balance between AI integration and traditional teaching methods Structured training and mentorship that offer hands-on experiences significantly boost teachers' confidence in utilizing ChatGPT, leading them to view the tool as advantageous for their instructional purposes.
In 2023, teachers face localized challenges in integrating AI tools like ChatGPT with curriculum goals and addressing ethical issues such as plagiarism prevention, highlighting the need for tailored professional development (Nguyen, 2023) While educators recognize ChatGPT's potential to personalize learning and alleviate workload, they also worry about its adverse effects on students' critical thinking skills This underscores the necessity of balancing AI-supported methods with traditional teaching practices (Dilzhan, 2024) Ultimately, teacher acceptance of ChatGPT is influenced by various contextual factors, necessitating well-designed professional development programs to facilitate effective implementation across diverse educational systems.
Teachers' acceptance of ChatGPT as a writing assistant is influenced by its perceived ease of use and usefulness The platform's accessibility, user-friendly design, and instant feedback contribute to its ease of use, while its ability to reduce workload, offer personalized feedback, and enhance creative teaching boosts its perceived usefulness These elements significantly motivate teachers to incorporate ChatGPT into their writing instruction Further exploration of teachers' acceptance is detailed in previous studies, which analyze the factors affecting the adoption of ChatGPT in classrooms and the variations in acceptance among teaching practices.
Teachers’ Application of ChatGPT
This section examines the integration of ChatGPT in teaching practices, highlighting its role in generating accurate content, lesson planning, grading student writing, designing assessments, and reviewing exercises By analyzing the benefits and challenges associated with each application, it offers a thorough overview of how ChatGPT enhances writing instruction and supports teachers in refining their teaching methods.
Teachers can benefit from using ChatGPT for lesson planning, as it efficiently structures lessons and enhances students' writing skills (Karaman & Gửksu, 2024) It also generates a variety of educational activities, making lessons more engaging (Baytak, 2024) However, there are limitations, such as potential inconsistencies where ChatGPT may combine outdated and modern teaching methods, necessitating careful review by educators (Dornburg & Davin, 2024; Park & Park, 2024) It is important to use ChatGPT to foster critical thinking rather than replace it (Van den Berg & du Plessis, 2023) Moreover, the tool is valuable for creating creative and personalized lesson plans and supplementary materials (Karakose & Tülübaş, 2023).
Grading student writing has gained attention with the use of ChatGPT as a writing assistant, as it can significantly save teachers time by managing tasks like grammar, spelling, and punctuation checks Research by Alnajashi (2023) indicates that ChatGPT-4 can accurately grade essays using specific rubrics, enabling educators to shift their focus from repetitive grading to more meaningful instruction However, despite its proficiency in structural aspects, ChatGPT faces challenges with nuanced elements such as creativity and tone, as noted by De Winter et al (2023).
Recent studies highlight the limitations of AI in grading open-ended tasks that demand complex assessment Flodén (2024) noted significant discrepancies between ChatGPT's scores and those given by human graders, particularly for questions needing specialized knowledge Additionally, Bui and Barrot (2024) found that essays scored by ChatGPT often fell short in depth compared to human feedback, underscoring the necessity for human oversight in the grading process.
Teachers are increasingly leveraging ChatGPT in educational settings for designing tests and creating review exercises, as noted by Nguyen (2024) This integration allows for the generation of a diverse array of tailored questions, facilitating efficient student assessment preparation and reinforcing knowledge Additionally, Linh (2023) highlights the advantages for teachers, including enhanced efficiency and personalization in test design, with ChatGPT providing quizzes, review materials, and suggestions for both formative and summative assessments The following section presents the theoretical framework of the study, detailing the key concepts and relationships explored.
The Theoretical Framework of the Study
The Technology Acceptance Model (TAM), introduced by Fred Davis in 1986 and published in 1989, aims to explain and predict user acceptance of new technologies Central to this model are two key constructs: perceived usefulness and perceived ease of use, which underscore the significant impact of a user's perception of technology's utility and usability on their intention to adopt and utilize it.
The original Technology Acceptance Model (TAM 1) has notable limitations, particularly regarding the Attitude Toward Using component, which does not consistently predict behavioral intention This is especially true when Perceived Usefulness (PU) exerts a stronger influence Research by Venkatesh and Davis (2000) indicates that in certain instances, PU directly impacts behavioral intention, diminishing the significance of Attitude Toward Using.
The researcher has adapted a simplified version of the original Technology Acceptance Model (TAM 1) introduced by Fred Davis in 1986, removing the Attitude Toward Using component to streamline the model while preserving the predictive power of Perceived Usefulness (PU) and Perceived Ease of Use (PEOU) on Intention to Use This theoretical framework supports the understanding of teachers' perceptions regarding the use of ChatGPT as a writing assistant and its application in educational practices The following sections will provide justifications for selecting this framework, along with an overview of the proposed TAM model in this study.
The simplified version of the Technology Acceptance Model (TAM) is chosen for this study due to its direct focus on how perceived ease of use and usefulness influence technology adoption This framework effectively motivates an exploration of English Language Teaching (ELT) teachers' acceptance of new technologies.
ChatGPT serves as an effective writing assistant, gaining recognition in educational research, particularly within language teaching and learning It provides a structured framework for evaluating the factors that influence the use of technology in educational settings, as highlighted by Venkatesh and Bala.
The study aims to identify key factors influencing teachers' adoption of AI in language teaching, focusing on perceived ease of use and perceived usefulness By incorporating both Intention to Use and Actual Use into the model, the research provides a comprehensive understanding of the journey from initial acceptance to actual implementation This approach aids in predicting the impact of AI tools on teaching practices and highlights potential barriers to their widespread adoption in English Language Teaching (ELT) contexts.
The adoption of technology, as outlined in this research study, occurs in three key stages Initially, external factors like system design influence individual perceptions of the technology, specifically regarding its ease of use (PEOU) and perceived benefits (PE) These perceptions subsequently shape users' intentions to utilize the technology, which ultimately translates into actual usage behavior The Technology Acceptance Model (TAM) underscores that user actions are primarily motivated by the perceived usefulness and ease of the technology, along with their intention to engage with it When individuals believe that the technology will yield positive results with minimal effort, they are more inclined to adopt it.
The proposed theoretical framework in this study centers on three key components influencing English Language Teaching (ELT) teachers' acceptance and utilization of ChatGPT as a writing assistant in general English courses First, it examines teachers' perceptions of ChatGPT's usefulness, particularly its potential to enhance student writing and teaching outcomes, alongside their views on its ease of use, which pertains to the simplicity of integrating ChatGPT into writing instruction Second, the framework addresses teachers' intention to use ChatGPT, reflecting their willingness and motivation to incorporate this tool in their writing classes This model illustrates a straightforward cause-and-effect relationship, indicating that teachers who recognize the benefits and ease of ChatGPT are more likely to implement it in their teaching practices.
22 see ChatGPT as useful and easy to use are more likely to adopt it, and this eventually leads to its actual use in the classroom
This section examines previous research on teachers' acceptance and use of ChatGPT in educational settings, aiming to pinpoint gaps in the existing literature that this study intends to fill.
Previous Studies
The increasing interest in AI tools such as ChatGPT in education emphasizes the necessity to investigate how English Language Teaching (ELT) educators integrate these technologies into their instructional methods Recent research has focused on two primary aspects: first, the acceptance of ChatGPT as a writing assistant by teachers in the classroom, as highlighted in studies by Mutammimah et al (2024), Strzelecki et al (2024), and Dehghani & Mashhadi (2024); and second, the practical application of these tools in teaching practices.
ChatGPT serves as a valuable writing assistant, as highlighted by various studies (Imran & Almusharraf, 2023; Mabuan, 2024; Song & Song, 2023) Despite previous research focusing on separate aspects of ChatGPT, there remains a significant gap in understanding how acceptance impacts its practical application This study aims to fill this gap by examining both the acceptance and utilization of ChatGPT among English Language Teaching (ELT) educators.
Perceived Ease of Use of ChatGPT as A
Intention to Use ChatGPT as a Writing Assistant
Actual Use of ChatGPT as a Writing Assistant Figure 2.2
The Theoretical Framework of the Study
2.6.1 Teachers’ Acceptance of ChatGPT in the Classroom
Research on teachers' acceptance of ChatGPT across various educational settings highlights key themes, including perceived ease of use, perceived usefulness, and external influences on its adoption These studies collectively provide important insights into teachers' perceptions of the tool and the factors that affect its integration into their teaching practices.
The ease with which ChatGPT can be used significantly impacts its acceptance among teachers, as evidenced in multiple contexts For example, Mutammimah et al
A recent study in Indonesia utilized the Technology Acceptance Model (TAM) framework to assess English teachers' views on ChatGPT, surveying 114 educators through an online questionnaire The analysis revealed that the perceived ease of use significantly influenced teachers' positive attitudes toward the platform, attributed to its user-friendly interface and minimal technical barriers that facilitated classroom integration without extensive training However, challenges emerged for teachers with limited digital skills, emphasizing the need for targeted professional development programs Similarly, Strzelecki et al (2024) found that a large sample of 629 university teachers in Poland appreciated ChatGPT's straightforward design, particularly for generating writing prompts and checking grammar Younger teachers favored the platform's simplicity more than older teachers, who showed reluctance due to unfamiliarity with AI technologies, indicating that demographic factors like age affect perceived ease of use In Vietnam, Nguyen (2023) reported that teachers valued ChatGPT's intuitive functionality for lesson planning and efficient feedback generation, though concerns about inadequate institutional support for training hindered some teachers' ability to fully leverage the platform.
24 the potential of the tool This suggests that while ease of use promotes acceptance, localized support mechanisms are crucial for effective adoption
The perceived usefulness of ChatGPT significantly influences teacher acceptance, as studies worldwide highlight its effectiveness in decreasing workloads and improving teaching quality.
In a 2023 study involving semi-structured interviews with 14 university teachers in Hong Kong, ChatGPT was highly valued for its capability to generate teaching materials and enhance lesson preparation efficiency Educators highlighted that the tool allowed them to concentrate on more advanced instructional activities, including facilitating discussions and offering personalized support to students Nonetheless, there were concerns regarding the potential overreliance on ChatGPT for routine tasks like grammar correction, underscoring the importance of maintaining a balanced approach to its usage.
A study conducted in 2024 surveyed 234 Iranian EFL teachers and found that ChatGPT significantly improved teaching efficiency by providing instant feedback on writing assignments and generating model answers, particularly aiding in correcting grammatical errors and enhancing writing organization However, ethical concerns arose regarding the potential misuse of AI-generated content by students Similarly, Nguyen (2023) in Vietnam reported that teachers viewed ChatGPT as beneficial for engaging students in writing activities, as it facilitated structured exercises and immediate feedback Despite these advantages, some educators expressed caution about excessive reliance on the tool, fearing it might impede students' critical thinking and creativity Overall, while both studies acknowledge the practical benefits of ChatGPT, they also highlight the importance of balancing its use to avoid overreliance.
Additionally, teacher acceptance of ChatGPT can be influenced by professional training programs, localized contextual needs, and the perceived balance
25 between AI integration and traditional pedagogical practices In Oman, Alrishan
(2023) examined the role of external factors in influencing ChatGPT adoption among
A study involving 280 pre-service EFL teachers revealed that structured training programs and mentorship greatly improved their confidence in utilizing ChatGPT for professional development Those who underwent hands-on training found the tool easier to use and more beneficial for their instructional requirements.
In recent studies across different countries, educators have recognized the potential of ChatGPT as a valuable tool for personalizing learning and reducing workload In Kazakhstan, Dilzhan (2024) found that teachers appreciated ChatGPT but were concerned about its impact on critical thinking skills, emphasizing the need to balance AI use with traditional teaching methods Similarly, a survey by Ogurlu and Mossholder (2023) in the U.S revealed that while teachers valued ChatGPT, the lack of comprehensive training limited its adoption, highlighting a strong demand for professional development to better integrate AI into teaching In Vietnam, Nguyen (2023) pointed out the necessity for localized training to help teachers navigate challenges in using ChatGPT for writing instruction, particularly in aligning it with curriculum goals and addressing ethical issues like plagiarism Additionally, Imran and Almusharraf (2023) conducted a systematic review of 30 articles on ChatGPT's role in higher education writing instruction, demonstrating its effectiveness in supporting drafting and providing feedback, while also calling for further research into teacher acceptance of AI tools.
26 specific pedagogical applications, particularly in understanding how varied educational contexts may influence the use of AI tools in English language courses
Studies reveal that teachers' acceptance of ChatGPT is influenced by perceived ease of use and usefulness, which are essential for fostering positive attitudes towards the tool Research by Mutammimah et al (2024) and Strzelecki et al (2024) highlights that ChatGPT's user-friendly design significantly enhances its acceptance among educators in Indonesia and Poland Furthermore, the perceived benefits, including workload reduction and improved instructional efficiency, were emphasized by Har (2023) in Hong Kong and Dehghani and Mashhadi (2024) in Iran However, the extent of adoption varies based on external factors like training and institutional support Alrishan (2023) in Oman and Ogurlu and Mossholder (2023) in the U.S found that structured training programs boost teachers' confidence in integrating ChatGPT, while Nguyen (2023) in Vietnam pointed out that a lack of professional development hinders adoption Despite global optimism regarding ChatGPT's potential, concerns about academic integrity and overreliance persist, as noted by Har (2023) and Dehghani and Mashhadi (2024), who emphasize the importance of careful implementation strategies to mitigate risks such as plagiarism and diminished critical thinking.
2.6.2 Teachers’ Application of ChatGPT in the Classroom
The integration of ChatGPT in educational settings demonstrates its ability to improve teaching methods, boost student performance, and tackle significant challenges in the learning process A comprehensive analysis of six studies showcases the diverse applications of ChatGPT by educators in the classroom.
Teachers have increasingly adopted ChatGPT to streamline instructional processes, particularly in writing education For instance, Imran and Almusharraf
A systematic review of 30 articles in 2023 highlighted the significant role of ChatGPT in higher education writing instruction, revealing that teachers utilize it for drafting, real-time feedback, and addressing language issues, thereby enhancing their workload management by automating tasks like grammar correction However, the study pointed out the variability in ChatGPT's application across different educational contexts, indicating a need for pedagogical frameworks to optimize its use Similarly, Mabuan (2024) conducted a mixed-methods study with 115 English Language Teaching (ELT) teachers in Metro Manila, finding that ChatGPT was integrated for lesson planning, grading, and generating model answers While teachers appreciated its ability to automate repetitive tasks, enabling a greater focus on interactive teaching, some expressed concerns about over-reliance on the tool for administrative duties, fearing it might diminish their engagement with students.
ChatGPT has been proven to enhance student writing skills, as demonstrated in a study by Song and Song (2023) involving 50 Chinese EFL learners over 12 weeks This research compared AI-assisted instruction with traditional teaching methods, where teachers utilized ChatGPT for immediate feedback on grammar, vocabulary, and essay structure The findings indicated significant improvements in writing proficiency among the AI-assisted group, alongside qualitative data showing increased learner motivation Educators recognized ChatGPT's effectiveness in scaffolding student learning, particularly in correcting common writing errors and promoting language development Similarly, Faiz et al (2023) explored ChatGPT's role in enhancing writing proficiency for 50 intermediate-level EFL learners in Pakistan.
A study involving 28 government colleges revealed that teachers used ChatGPT to offer focused feedback on grammar, vocabulary, and composition, leading to more efficient writing instruction and improved student writing quality However, concerns about plagiarism emerged, highlighting the necessity for clear guidelines on responsible AI use in educational settings This research underscored ChatGPT's dual role as both a valuable instructional aid and a potential ethical challenge.
ChatGPT has been shown to enhance student autonomy in writing, particularly in EFL classrooms, as explored by Tseng and Lin (2024) Their research identified three primary applications: grammar correction, essay structuring, and guided writing prompts, emphasizing the importance of teaching students to effectively interact with ChatGPT for self-assessment and reflective practices This shift from teacher-led instruction to student-centered learning encourages learners to take responsibility for their language development Ahn et al (2024) further examined the use of ChatGPT in ELT settings, where teachers utilized it for generating writing prompts, providing grammar feedback, and assisting in revisions Their findings indicated that ChatGPT not only streamlined the feedback process but also engaged students in reflective writing tasks Nonetheless, teachers highlighted the need to balance ChatGPT's use with traditional methods to preserve critical thinking and originality in student writing.
Summary of the Chapter
The literature review reveals the evolution of AI in education and emphasizes ChatGPT's role as a writing assistant It highlights significant benefits of writing instruction while identifying research gaps in Vietnamese English Language Teaching (ELT), particularly regarding the acceptance and use of ChatGPT This context sets the stage for exploring its application in general English courses.
METHODOLOGY
Research Design
This study investigates teachers’ perceptions of ChatGPT as a writing assistant, focusing on their acceptance of the tool and its application in general English courses Utilizing a qualitative research design within the interpretivism paradigm, the research aims to capture the nuanced experiences of English Language Teaching (ELT) educators According to Creswell (2014), qualitative research explores and understands the meanings individuals or groups assign to social issues, involving emerging questions, data collection in participants’ settings, and inductive analysis that builds from specific observations to broader themes.
To establish a theoretical foundation for the qualitative research design employed in this study, the interpretivism paradigm was adopted According to Gephart et al
Interpretivism, as highlighted by Nickerson & Mcleod (2023) and Pham (2018), focuses on understanding the subjective meanings of human actions within their social contexts Unlike positivism, which seeks objective truths through quantitative data, interpretivism examines phenomena in natural settings, prioritizing participants’ perspectives and the complexities of human behavior This qualitative approach values depth, context, and rich data, aligning perfectly with the objectives of this research.
The study investigates the diverse perceptions and experiences of English Language Teaching (ELT) teachers regarding their use of ChatGPT as a writing assistant Adopting an interpretivist approach, it highlights the significance of participants' perspectives, underscoring how knowledge is constructed through interpretation and the impact of contextual factors on their opinions.
This study employs qualitative methods to explore English Language Teaching (ELT) teachers' acceptance and use of ChatGPT as a writing assistant in general English courses By focusing on teachers' perceptions and experiences, the research aims to provide a comprehensive and nuanced understanding that qualitative approaches uniquely facilitate.
Qualitative research facilitates a thorough exploration of participants' perspectives within their natural environments, making it particularly suitable for investigating the experiences and perceptions of teachers using ChatGPT as a writing assistant in general English courses The study's first research question focuses on capturing the nuanced attitudes of teachers through rich, descriptive data, while the second question examines the varied and context-specific integration of ChatGPT into teaching practices This approach is essential for uncovering complexities that quantitative methods may overlook, as noted by Creswell & Poth (2018) Furthermore, the inclusion of diverse participants with varying teaching backgrounds allows for a detailed examination of unique insights and variations in the acceptance and application of ChatGPT, which is crucial for understanding the intricate realities of technology integration in English Language Teaching (Patton, 2002).
Qualitative methods, including interviews and thematic analysis, are valuable for exploring individual experiences and narratives, as they offer a deep understanding
This study utilized semi-structured interviews to gather in-depth insights into teachers' perceptions of ChatGPT as a writing assistant By exploring their experiences with this AI tool, the research highlighted how it influenced their teaching practices and acceptance of artificial intelligence in the classroom, effectively capturing the complexities of human behavior and social trends.
In this research, manual coding techniques were utilized for data analysis to systematically interpret information, employing codes and categories to organize findings and create a narrative about teachers’ attitudes and experiences (Creswell & Creswell, 2022) This approach provided a comprehensive understanding of the implications of ChatGPT in educational contexts Following the data analysis, the results from semi-structured interviews regarding teachers’ acceptance and use of ChatGPT as a writing assistant were interpreted independently.
This study utilized an interpretivist approach and qualitative methods, specifically semi-structured interviews and thematic analysis, to investigate teachers' acceptance and use of ChatGPT as a writing assistant in general English courses.
Research Context
The research for this study was conducted at College A, located in Ho Chi Minh City This private institution offers three modes of delivery: on-campus, in-service, and distance education Over the past two decades, College A has established itself as a reputable provider of skilled human resources.
Every year, the English Faculty welcomes more than 300 non-English majors into a 2.5-year program designed for 12th-grade graduates who meet specific academic criteria To enroll, students must complete a placement test that determines their placement in one of three general English classes.
“Four Corners” (Richards & Bohlke, 2013), is used throughout Teachers provide
To ensure effective teaching and compliance with regulations, 35 course outlines, lesson plans, and handbooks were prepared in advance In the second semester of 2023-2024, approximately 800 non-English majors participated in twenty classes tailored to CEFR B1 standards, all of which were enhanced with projectors, speakers, and electronic learning devices.
In English non-major courses, students typically engage in about 45 class periods per semester, with writing instruction comprising approximately 15 of those periods, equating to 45 minutes of weekly writing practice Educators focus on fundamental writing skills through assignments like short essays, formal letters, and reports They often combine reading and writing tasks, prompting students to critically analyze texts and utilize them as writing models Additionally, teachers facilitate learning through guided exercises and peer reviews, aiding students in paragraph construction, idea organization, and the application of correct grammar and vocabulary.
Participants and Sampling Strategies
The current study focused on a target population of 14 English teachers, including both visiting and official lecturers, with 6 teachers actively participating This selection underscores the significance of sampling, as highlighted by Morrison (1993) in Cohen et al (2007), who asserts that research quality hinges not only on the chosen methodology and data collection tools but also on the effectiveness of the sampling strategy.
“selecting units of analysis (e.g., people, groups, artifacts, settings)” to enhance the researcher’s ability to answer research questions (Tashakkori & Teddlie, 2003, p
In qualitative research, various sampling methods exist, such as censuses, non-probability, and probability sampling (Guest et al., 2013, p 78) However, for studies aiming to gather rich, in-depth information without seeking generalizations, censuses and probability sampling are unsuitable Consequently, this study opted for non-probability sampling, specifically utilizing opportunity, accidental, or convenience sampling to collect qualitative data.
In this study, convenience sampling was employed as a qualitative strategy to select participants, specifically English lecturers at College A who were accessible and willing to participate This method is recognized for its practicality and ease of use in research (Onwuegbuzie & Collins, 2007; Nguyen, 2013) Participants were chosen based on three criteria: they had to be English lecturers (both visiting and official), have experience using ChatGPT for writing instruction in general English courses, and be willing to engage in the study This approach ensured that the selected participants possessed relevant teaching experience and familiarity with ChatGPT, facilitating targeted insights into its application in English writing instruction.
In qualitative research, focusing on a smaller group allows for a deeper understanding of individual circumstances, with the goal of generalizing findings to theory rather than to populations (Bryman, 2012) For this study, 6 out of 14 teachers were selected for semi-structured interviews based on their convenient access to the site Participants were excluded if they lacked experience with ChatGPT in writing instruction, were unavailable or unwilling to participate, or withdrew due to teaching workloads, ensuring that the sample consisted solely of individuals with relevant insights.
Table 1 outlines the profiles of six interviewed teachers, revealing that the majority, four out of six, are male The ages of these educators span from twenty-five to forty-two, with 60% falling within the 30-40 age bracket Their teaching experience ranges from two to eighteen years, and notably, 60% of the teachers have between five and twelve years of experience.
With 37 years of combined experience, the teaching staff includes four educators with a Master's degree and two with a Bachelor's degree Most teachers demonstrate moderate to high familiarity with technology, with five having received training in AI tools Only one teacher reported no participation in AI training, while the others have successfully integrated various techniques into their teaching practices using ChatGPT.
A Summary of Participants’ Information in the Semi-structured Interviews
T01 Female 25 2 yrs BA Moderate Yes
T02 Male 35 10 yrs MA High Yes
T03 Male 28 5 yrs BA High Yes
T04 Female 42 18 yrs MA Low No
T05 Male 39 12 yrs MA Moderate Yes
T06 Male 30 8 yrs MA Moderate Yes
Data Collection Instruments
This study utilized qualitative semi-structured interviews with six volunteer teachers from College A to collect data This section elaborates on the data collection methods and the reasoning for their selection.
A research instrument is defined as a tool for data collection, chosen according to the research objectives, data type, sample, and design (Kothari, 2004) Interviews are particularly effective for gathering qualitative data (Kajornboon, 2005).
Research fosters the exchange of ideas, emphasizing the importance of social interaction in knowledge creation (Brinkmann & Kvale, 2018) Additionally, conducting interviews enables researchers to utilize effective communication skills, thereby enriching the exploration of participants' perspectives and ultimately enhancing the validity and reliability of the research findings (Gray).
2021) while collecting detailed data and addressing ambiguities
Bell and Waters (2018) categorize interviews into three types: structured, semi-structured, and unstructured Structured interviews use a fixed set of questions, which may restrict the depth of understanding regarding participants' attitudes In contrast, unstructured interviews offer a more conversational approach, enabling researchers to adapt questions for richer data collection; however, they can complicate the exploration of specific themes under time constraints, potentially impacting data validity (Ayres, 2012; Carson et al., 2011; Cohen et al., 2018) Semi-structured interviews strike a balance by integrating a guided question framework with the flexibility to delve deeper into insights while efficiently covering essential topics.
Also, observation is a widely recognized approach in qualitative research for understanding behaviors and practices in naturalistic settings (Creswell & Poth,
In exploring the use of ChatGPT by English Language Teaching (ELT) teachers, observation was initially considered as a research method due to its strength in capturing real-time interactions and contextual nuances (Creswell & Poth, 2018) However, the challenges of effectively observing ChatGPT usage, which often occurs in diverse contexts and outside of classroom settings, render this method impractical Consequently, relying solely on observation would fail to adequately capture the full range of teachers' practices and perceptions regarding ChatGPT in their teaching methodologies.
In this study, the researcher chose semi-structured interviews due to their balance of structure and flexibility, enabling an in-depth exploration of English Language Teaching (ELT) teachers' acceptance and use of ChatGPT This method allows for the adjustment of questions based on participants' responses, aligning with specific research questions and target populations Additionally, semi-structured interviews facilitate the emergence of new ideas and themes, enriching the data collected Importantly, this approach addresses sensitive ethical considerations, allowing researchers to ask deeper questions and gather comprehensive information Participants share detailed experiences with ChatGPT as a writing assistant in general English courses, fostering a thorough understanding of the topic.
Creswell and Creswell (2018) highlight various methods for conducting interviews, including face-to-face, phone, online, or focus group discussions In this study, researchers held online interviews with participants via Microsoft Teams, each lasting approximately 60 minutes The convenience of online interviews allowed for flexible scheduling, making it easier to accommodate participants' availability Conducted in Vietnamese with a semi-formal approach, these interviews fostered a comfortable atmosphere that encouraged smooth and uninterrupted conversations.
This approach greatly improved the likelihood of obtaining detailed and rich information The semi-structured nature of the interviews allowed the researcher to adapt and rephrase questions as necessary, ensuring a smooth and coherent dialogue while still gathering systematic and comprehensive data relevant to the research questions (Minichiello et al., 2008, p 52) Additionally, the researcher took note of significant points during the interviews.
In this research study, six interviews were conducted with English Language Teaching (ELT) teachers to gain an in-depth understanding of their perspectives Spacing the interviews over a period of three weeks, with two interviews each in the first two weeks and the final two in the third week, proved to be a practical approach This scheduling allowed for adequate time for transcription, reflection, and necessary adjustments based on emerging insights.
The interview questions, totaling nineteen, were crafted using components of the Technology Acceptance Model (TAM) as outlined in Appendix 4 Following the guidance of Deslonde and Becerra (2018), the researcher utilized a survey questionnaire to formulate these interview questions, qualitatively adapting and developing them from items originally proposed by Davis.
In a study exploring the acceptance of ChatGPT as a writing assistant among English Language Teaching (ELT) teachers, the research utilized a structured interview format based on the Technology Acceptance Model (TAM) The interview comprised two main sections: background information and semi-structured questions focusing on technology acceptance factors The background section included three questions to establish rapport and comfort with the participants, while the acceptance section featured sixteen questions divided into four dimensions: perceived ease of use, perceived usefulness, intention to use, and actual use The researcher maintained a minimal speaking role, employing probing techniques to elicit deeper insights, thereby facilitating a comprehensive understanding of teachers' perspectives on integrating ChatGPT into their teaching practices At the conclusion of each interview, the researcher expressed gratitude to the participants for their contributions.
Validity is the assurance that a measurement tool accurately assesses what it intends to measure, ensuring that the data generated reflects the intended traits or concepts (Winter, 2000, as cited in Cohen et al., 2018) Additionally, validity encompasses the degree to which interpretations derived from the data are backed by relevant theories and supporting evidence (Ary et al., 2002, p 267).
This study enhanced data precision and completeness through the use of audio and video recordings, along with thorough transcription, translation, and coding The verbatim transcripts from semi-structured interviews were meticulously reviewed for accuracy by two EFL experts from College A, ensuring the credibility of the data These transcripts were translated into English and shared with the six interviewees for clarification, ensuring their narratives were accurately represented To further validate the data, the researcher coded the transcribed excerpts twice Following the transcription phase, the researcher piloted the interview questions with participants similar to the target group, adhering to a structured five-step process to enhance question validity Participants were contacted via email to explain the pilot's purpose, and after providing clear instructions, the interviews were conducted The semi-structured format allowed for adaptive questioning based on emerging information, and two pilot tests were carried out to refine the interview questions, with a recommended gap of several days to a week between sessions for optimal feedback.
After conducting Pilot Test 1 on May 1st and incorporating feedback for Pilot Test 2 on May 6th, we refined the interview questions to ensure iterative improvement This method allows for effective analysis and enhancement of the testing process without significant delays.
Following a pilot interview with two English Language Teaching (ELT) teachers, adjustments were made to enhance the clarity and flow of the questions One teacher pointed out that the phrase "technology integration" in the question regarding ChatGPT usage was too vague, prompting the inclusion of a brief explanation to clarify that it encompasses the use of digital tools or software in classroom activities Additionally, another teacher observed that certain follow-up questions needed refinement for better comprehension.
“How has ChatGPT affected the workflow of your daily writing instruction tasks?” were repetitive, which led to merging related questions to streamline the guide.
Data Collection Procedures
This study utilized qualitative semi-structured interviews with teachers who voluntarily agreed to participate, detailing the data collection procedures involved Further information on these processes is provided below.
Data collection is a systematic process aimed at gathering in-depth information from research participants, leading to the creation of language data and discourses (Polkinghorne, 2005) To ensure accuracy and consistency, the data collection procedure was meticulously followed This involved a four-step process for gathering data through semi-structured interviews.
Step 1 Choosing the participants for the interview process
After securing ethical approval, the researcher worked with the board of College A to start the research and sent recruitment invitations, including consent
43 forms in English (see Appendix 2) The purpose and significance of the study were clearly communicated to ensure full cooperation
Step 2 Scheduling interviews with participants
The researcher conducted classroom visits to verify the participation of teachers who agreed to allow the use of ChatGPT in their writing classes After outlining the study's objectives, potential risks, and the participants' right to withdraw without repercussions, signed consent forms were collected from the teachers Subsequently, the researcher reached out to participants via email to schedule interviews.
In these emails, the researcher clarified the interview objectives and provided a detailed overview of the procedures
The researcher conducted interviews with participants according to the established schedule and guidelines, encouraging honesty and open communication To facilitate a more natural expression of ideas, participants were allowed to respond in Vietnamese, effectively minimizing language barriers and enhancing the depth of their responses.
The author initiated each interview by obtaining participants' consent to record the session Once permission was granted, participants completed a warm-up task, responding to simple questions like, “How was your teaching today?” and “Have you had dinner yet?” This approach aimed to create a relaxed atmosphere and familiarize participants with the interview process before delving into the main questions Throughout the main interview, participants were encouraged to elaborate on their answers, with the author using follow-up prompts such as, “Could you tell me more about that?” to seek clarification and deeper insights.
44 me more about ?” or “Would you explain that in more detail?” In cases where responses appeared unclear or off-topic, additional questions were asked, such as,
The study aimed to evaluate the effectiveness of new teaching tools compared to traditional methods Interviews were held in a calm, distraction-free setting, allowing participants to articulate their thoughts thoroughly Each session concluded with the researcher thanking the participants for their cooperation before ending the recording.
Step 4 Receiving confirmation of the transcripts
The recordings of the interviews were transcribed and then sent to the participants for review, allowing them to verify the content or suggest any necessary changes.
Data Analysis
Effective data coding is essential for researchers as it streamlines various tasks, including data organization, differentiation of items and types, and sorting or filtering processes (Guest et al., 2013) To enhance data quality and facilitate analysis, it is crucial to standardize transcription and translation Guest et al (2013) recommend incorporating at least three key identification elements in the coding process: source ID, data labels, and file names Examples of file names include digital audio or video recordings, scanned consent documents, and transcripts.
In the current study, file names were found to be ineffective for identifying codes, leading the researcher to opt against their use for data coding Instead, the researcher focused on patterns to create codes, as patterns fundamentally reflect the relationship between unity and multiplicity (Stenner, 2014, p 136, as cited in Saldaña, 2016).
The qualitative data in this study were categorized based on three key elements: source ID, data labels, and patterns Interview records were labeled using a simplified syntax that consists solely of the participant number.
To ensure clarity and efficient organization of research data, a streamlined naming convention is employed, where identifiers like “01_Interview” denote the recording of Participant 01’s interview and “02_Transcript” indicates the transcript of Participant 02’s session Additional details, such as the research phase, can be included, for example, “03_Final” for the final transcript of Participant 03, minimizing unnecessary categorization.
In this study, manual coding was employed to preserve essential information, as recommended by Wahyuni (2013) The coding process commenced with open coding, which focused on identifying, naming, categorizing, and describing actions or events by breaking down data into manageable segments for analysis (Nguyen, 2013) Following this, axial coding was utilized to connect themes and develop a more comprehensive understanding of the data During this phase, relevant data were systematically organized into themes and subthemes aligned with the research objectives The subsequent section elaborates on this stage in the current study.
Thematic analysis, as outlined by Braun and Clarke (2006), was utilized to analyze and interpret interview data, providing qualitative insights into teachers' perspectives This method focuses on identifying, analyzing, and reporting patterns or themes within the data, offering researchers flexibility and accessibility in their analysis.
The thematic analysis conducted in this study focused on examining qualitative data to uncover recurring patterns linked to pre-defined themes that guided the creation of semi-structured interview questions The analysis aimed to identify themes and subthemes associated with factors believed to influence teachers' experiences.
46 inductive approach to thematic analysis was also employed with the hope to identify possible new themes
This research study utilized only confirmed transcripts for analysis, which were translated from Vietnamese to English by the researcher and an ELT translator following mutual agreement The thematic analysis was conducted using the five-step process established by Braun and Clarke (2006) and updated in 2019, a method recognized in previous research (e.g., Alshibany, 2018; Thomson, 2016) The steps included familiarizing with the data, generating initial codes, searching for themes, redefining and naming themes, and producing the final report.
In this stage, the researcher engaged in a thorough reflection of the data by coding and identifying themes, following the methodology outlined by Howitt (2013) The semi-structured interview data were transcribed verbatim and organized into individual folders for each teacher, with careful accuracy checks during the transcription process To facilitate later analysis, the transcripts were translated into English The researcher meticulously reviewed the entire dataset multiple times to uncover meanings, patterns, and insights, emphasizing key language segments that suggested potential themes Although this process is time-consuming, it is essential for achieving a comprehensive understanding of the content's breadth and depth, as supported by Braun and Clarke (2006).
Coding is the initial step in theme generation, utilized to condense extensive data into smaller, analyzable units (Denscombe, 2010) This process entails segmenting data into manageable components for effective analysis.
In 2010, Howitt defined line-by-line coding as “brief descriptions of small chunks of data” (2013, p 176) The researcher analyzed the transcripts and developed codes that corresponded with predetermined themes, marking this process as a formal step in data analysis.
In thematic analysis, Braun and Clarke (2006) propose two coding methods: the data-led approach, which derives codes directly from the data, and the theory-led approach, which utilizes a theoretical framework to shape initial coding In my research, I employed both methods, using the Technology Acceptance Model (TAM) as the theoretical framework to guide the design of semi-structured interview questions and the coding process The analysis focused on TAM's core constructs, such as perceived usefulness, perceived ease of use, intention to use, and actual use Coding and interpretation were aligned with predefined theoretical categories, leading to the identification of consistent patterns in the interview data, which were categorized into subthemes like individual learning paths and error detection, ultimately organized under main themes such as personalized learning and instant feedback.
During this stage, the researcher analyzed the identified codes by highlighting their similarities and differences Relevant data segments were organized and categorized according to predefined themes, which were subsequently refined and developed The researcher further enhanced these themes and their emerging sub-themes based on the analyzed code segments, as illustrated in Table 2.
A Sample of Searching Themes from the Interview Data
Excerpts Themes and sub-themes
“ChatGPT provided instant feedback, which made my job much easier and saved me a lot of time […]
Ease of Access and Use
Before utilizing ChatGPT, I dedicated countless hours to manually editing students' writing Now, with the assistance of ChatGPT, I can automate much of the tedious editing process, enabling me to focus more on providing in-depth, higher-level feedback.
Reduce teaching workload in editing
The researcher ensured that each theme was distinctly defined and separated to prevent repetition, facilitating a more thorough analysis Regular checks on the data and coding were conducted to ensure alignment with the established themes (Howitt & Cramer, 2014).
Ethical Considerations
In academic research, ethics refers to how appropriately the researcher behaves, the rights of the participants, and how the study might affect them (Saunders et al.,
In compliance with ethical guidelines, the study implemented several protective measures for participants, including voluntary participation, the right to withdraw without penalty, informed consent, and the assurance of anonymity and confidentiality (Creswell, 2013; Mertens, 2015; Cohen et al., 2018) Adhering to the principles outlined by Padgett (1998) and Fraenkel et al (2018), the research emphasized privacy, honesty, and the prevention of harm Prior to the interviews, the researcher ensured that teachers were well-informed about the research aims and processes, fostering a comfortable environment for participation.
To uphold anonymity and confidentiality, teachers' identities were omitted, and participants were assigned codes (e.g., T01 for the first teacher) The researcher ensured transparency and honesty by delivering accurate information and clarifying the interview process Data was securely stored to protect confidentiality, and efforts were made to highlight the research's significance in the English Language Teaching (ELT) field By adhering to these ethical standards, the study aimed to provide valuable insights while prioritizing the privacy and well-being of participants.
In literature, "trustworthiness" is often synonymous with "authenticity" or "credibility," highlighting the importance of accuracy in research findings from the perspectives of researchers, participants, and readers (Creswell).
& Miller, 2000) In this study, four strategies were employed to ensure the accuracy of the findings, as suggested by Patton (2002, as cited in Li, 2017) and Creswell
In the study conducted in 2018, the triangulation method was utilized, as outlined by Patton (2002) and cited in Li (2017) This technique involved the integration of various research tools, including semi-structured interviews, thematic analysis, and manual coding, to gather data relevant to the two research questions The data collected from these diverse methods were subsequently compared and contrasted, enhancing the accuracy and reliability of the findings.
The study utilized semi-structured interviews to collect data, followed by member-checking to ensure the accuracy of findings, as recommended by Creswell (2018) After verbatim transcription of the interviews, key codes and themes emerged, which participants reviewed for accuracy Translated quotes were verified by both the researcher and an invited lecturer, with all participants agreeing to the final versions To enhance the richness of the findings, a detailed description of the setting was provided, making the results more realistic Additionally, to minimize bias, the researcher carefully crafted neutral interview questions that balanced positive and negative perspectives and employed open-ended formats to encourage comprehensive responses from participants.
Chapter Summary
This chapter presents the methodology for examining teachers' acceptance and use of ChatGPT as a writing assistant in general English courses Utilizing an interpretivist paradigm and qualitative approach, the study captures in-depth insights into teachers' experiences Data was collected through semi-structured interviews and analyzed using thematic coding Additionally, the chapter outlines the research context, participant selection, sampling methods, and ethical considerations, ensuring the study's validity and trustworthiness.