English major students’ perceptions towards using the chat gpt ielts writing correction tool for improving ielts writing skills

BANKING ACADEMY FACULTY OF FOREIGN LANGUAGES GRADUATION THESIS ENGLISH-MAJOR STUDENTS’ PERCEPTIONS TOWARDS USING THECHAT GPT IELTS WRITING CORRECTION TOOL FOR IMPROVING IELTS WRITING SKI

INTRODUCTION

Background of the study

Effective knowledge acquisition methods in academia are essential for optimizing time and effort in the teaching and learning process, particularly with the rise of artificial intelligence (AI) technologies One notable application is the use of Automated Essay Scoring (AES) systems to enhance IELTS writing skills among English as a Foreign Language (EFL) students These AI tools, including the Chat GPT IELTS Writing Correction Tool, provide valuable support by grading essays and outlining specific formats, thereby improving writing proficiency while saving time Research indicates a growing acceptance of AES tools in academic writing, with studies highlighting their effectiveness in enhancing learners' skills However, some researchers express concerns about the potential drawbacks of these technologies This research aims to explore EFL students' perceptions of the Chat GPT IELTS Writing Correction Tool, filling a gap in understanding its impact on their learning experience and confirming their expectations of its effectiveness as a supportive educational resource.

Research aim and research questions

This study investigates the perceptions of English-Major students at the Banking Academy regarding the enhancement of their IELTS writing skills through the Chat GPT IELTS Writing Correction Tool Additionally, the findings serve as a reflection on the attitudes of these students towards utilizing an Automated Essay Scoring (AES) tool for improving their IELTS writing proficiency.

This research attempt to answer these following questions:

RQ1: What are English-Major students’ perceptions towards the Chat GPT IELTS Writing Correction Tool?

Significance of the study

This research is crucial for stakeholders aiming to enhance academic writing skills, particularly through the Chat GPT IELTS Writing Correction Tool This tool offers valuable support for English language students preparing for the IELTS exam, helping them improve their study efficiency and scores Educators and instructors in IELTS training can also benefit by integrating this innovative tool into their teaching methods to better engage learners Furthermore, the findings of this research will be instrumental for future studies, especially in light of the rapid advancements in artificial intelligence technology.

Scope and limitation of the study

Due to time constraints, the research will primarily focus on a group of English language major students at the Banking Academy.

Research structure

Chapter 1 outline the rationale for choosing the subject, as well as the study's aims, scopes and limitations, and structure

Chapter 2 offers a review of related literature on the concept of AES as well as the definition of the Chat GPT IELTS Writing Correction Tool, and how this tool is valued in the area of learning IELTS Writing Skills

Chapter 3 provides an explanation of the research technique In terms of study locations, research design, participants, data collection, research instrument, and statistical treatment

Chapter 4 discusses the findings and results of the research

Chapter 5 is the final chapter of the research, which provides a brief summary and many recommendations for more research.

REVIEW OF RELATED LITERATURE

The Technology Acceptance Model (TAM), developed by Davis in the 1980s and rooted in the Theory of Reasoned Action, is a framework for predicting user behavior towards technology It emphasizes that technology acceptance involves individuals or groups evaluating and experiencing new technology's features and benefits According to Davis, two key factors influence technology acceptance: Perceived Ease of Use and Perceived Usefulness Perceived Ease of Use refers to how easy users find the technology to navigate and utilize, while Perceived Usefulness is the degree to which users believe the technology enhances their performance in tasks High technology acceptance occurs when users recognize that the technology saves time, reduces costs, and boosts work efficiency.

Beyond the primary factors of the Technology Acceptance Model (TAM), additional elements such as beliefs, attitudes, intention to use, actual usage, and user expectations play a crucial role in enhancing our understanding of technology acceptance These variants contribute to more precise predictions of user behavior regarding technology adoption.

Numerous studies have applied the Technology Acceptance Model (TAM), developed by Davis in 1980, to various research processes, demonstrating its effectiveness in predicting user acceptance of technologies This model has been utilized in contexts such as wireless LAN (Yoon & Kim, 2007), mobile learning (Park, Nam & Cha, 2011; Tai & Ting, 2011), e-learning (Ong & Lai, 2006; Roca & Gagné, 2008), PDAs (Arning & Ziefle, 2007), and online communities (Liu et al., 2010).

While existing studies primarily examine quality, value, and convenience, there is a notable lack of research on student satisfaction regarding these tools This study aims to address this gap by emphasizing the importance of expectation confirmation in understanding student satisfaction levels.

Artificial intelligence (AI) technologies significantly enhance content writing by providing valuable suggestions and prompts for idea generation One notable development is Automated Essay Scoring (AES), which uses computer algorithms to evaluate and score essays without human intervention AES combines scoring algorithms with software that aids learners in error correction and generates essay prompts, allowing for simultaneous analysis and evaluation of user-written content This system is recognized in education as an effective and time-saving tool for essay assessment, particularly in exams and online courses Despite its strengths and weaknesses, AES is a valuable resource for improving the academic writing skills of English learners Over the past 50 years, AES has garnered considerable attention from practitioners and researchers Various software systems, such as Project Essay Grade (PEG), have been developed based on AES; however, PEG has faced criticism for focusing primarily on surface structures while neglecting content-related aspects.

In the 1990s, numerous Automated Essay Scoring (AES) systems were developed to enhance the essay grading process and mitigate controversies similar to those surrounding PEG Notably, IntelliMetric, created by Vantage Learning, stands out as an AES system that evaluates various factors in an essay, delivering immediate scores and constructive feedback on writing quality This innovative learning engine leverages the collective expertise of experienced human raters through a blend of artificial intelligence (AI), natural language processing (NLP), and statistical methods.

IntelliMetric employs advanced natural language processing techniques and a multi-criteria evaluation model to assess writing quality and performance, earning favorable feedback from the academic writing community It is widely utilized in the Writing sections of prominent exams, including the Test of English as a Foreign Language (TOEFL), Scholastic Assessment Test (SAT), and American College Testing (ACT).

The importance of succeeding in international English language tests has grown due to increased foreign migration to English-speaking countries for work and education Proficiency assessment exams in English enable learners to establish specific language learning goals and evaluate their skills in both everyday and advanced academic contexts The International English Language Testing System (IELTS) has gained significant popularity, especially among Vietnamese students majoring in English, reflecting its substantial development since its inception in 1980 IELTS includes two core exams—speaking and listening—and two specialized tests—writing and reading—and is now widely recognized as a prerequisite for courses in English-speaking nations.

1989 Holden and Taylor, 1997) IELTS offers a single exam that may be used to quantify a person's English competence independent of prior knowledge or experience Pearson, William S (2018)

The IELTS exam evaluates English proficiency on a scale from 0-9, corresponding to levels B1 to C1 of the Common European Framework of Reference for Languages (CEFR) It consists of two main types: IELTS Academic, designed for students aiming to study abroad, and IELTS General Training, which focuses on real-life situations for those pursuing migration or vocational training Notably, up to 77% of IELTS Academic test-takers also undertake proficiency assessments (IELTS, n.d.) This study specifically examines English language major students at the Banking Academy, concentrating on the IELTS Academic test format.

The IELTS Academic Writing test is structured to meet rigorous academic standards for students applying for scholarships or higher education institutions It comprises two tasks: Task 1 requires candidates to summarize information from a chart, diagram, or process, while Task 2 involves writing an essay that showcases logical reasoning and advanced academic language to articulate opinions and support arguments with evidence Candidates have a total of 60 minutes to complete both tasks, with recommendations suggesting that approximately 20 minutes should be allocated to Task 1 for optimal performance.

20 minutes, while Task 2 should take approximately 40 minutes

Task 2 of the IELTS Writing test typically focuses on academic subjects relevant to university or postgraduate studies, as well as everyday topics related to education, society, and global issues To excel in the IELTS, students must possess a foundational understanding of general social issues (Wilson, 2010).

IELTS examiners, trained and certified, evaluate writing tasks separately, with Task 2 carrying a higher scoring weight than Task 1 (Uysal, 2009) Candidates must exhibit advanced critical thinking and language skills in Task 2, requiring them to logically organize their ideas and provide relevant real-life evidence to support their arguments Additionally, Task 2 demands a richer vocabulary and more varied grammar structures to effectively convey ideas clearly and succinctly.

The IELTS Writing section uses a scoring system that ranges from band 0 to 9, with each score calculated by averaging the scores from various skills and rounding to the nearest half or whole number According to Uysal (2009), IELTS examiners assess writing performance based on specific band descriptors While the exact process for converting these criteria into individual scores remains unclear, the evaluation focuses on four key areas: task attainment (TA), coherence and cohesion (CC), lexical resource (LR), and grammatical range and accuracy (GRA) (British Council, n.d.; UCLES, 2004).

The scoring criteria for the writing test range from 0 to 9, indicating different proficiency levels that candidates must understand to maximize their performance (Pearson, 2018) Each criterion significantly influences the final score, making it essential for candidates to address all task requirements through logical arguments and accurate data for Task Achievement Coherence and Cohesion (CC) assess the overall structure of the essay, ensuring it is clear and effectively presents main and supporting arguments The Lexical Resource (LR) criterion evaluates the candidate's vocabulary diversity and spelling accuracy, while the Grammatical Range and Accuracy (GRA) criterion measures the candidate's ability to use various grammatical structures effectively Understanding these criteria is vital for candidates aiming to improve their writing skills and achieve better scores.

The clear establishment and development of assessment criteria for the IELTS Writing test have enabled learners to effectively identify their learning objectives and pinpoint challenges in their learning journey This study will specifically address one of the four assessment criteria outlined above.

RESEARCH METHODOLOGY

Research locale

This research was conducted among English language major students at the Faculty of Foreign Languages of Banking Academy.

Research design

This study utilizes a mixed-methods sequential explanatory research design, integrating both qualitative and quantitative approaches For the quantitative aspect, data is collected using the Likert scale via Google Forms, while qualitative insights are gathered through semi-structured interviews By combining these methods, the research aims to deliver a comprehensive understanding of the issue at hand, ultimately leading to rich conclusions and actionable solutions As highlighted by Pardede (2019), employing a mixed-methods sequential explanatory design enhances the accuracy and depth of research findings, enabling researchers to gain a clearer perspective on the challenges they are exploring.

Participants

This study involved 46 participants from the English Language Department who completed an online survey regarding the Chat GPT IELTS Writing Correction Tool Given the tool's novelty and the growing popularity of IELTS exam preparation among diverse age groups, the survey aimed to include all students who had previously utilized this tool in their writing preparation To ensure data reliability, only those who had experience with the Chat GPT IELTS Writing Correction Tool were included, while others who did not meet the criteria were excluded All participants volunteered for the study, with assurances that their personal information would remain confidential and used solely for research purposes.

The GPT IELTS Writing Correction Tool was chosen for the study due to its relevance for English language students at the Banking Academy, who are potential users preparing for the IELTS exam to advance their career goals Furthermore, all survey participants must possess a strong foundational knowledge of the language.

Research instrument

The mixed methods approach effectively combines quantitative and qualitative techniques to gather and analyze data, offering a thorough understanding of English language students' perceptions of the Chat GPT IELTS Writing Correction Tool in their writing studies This method enhances the reliability and accuracy of research findings by leveraging the strengths of both approaches Inspired by Zamzam Bibi's (2024) related research, I believe that this approach is the most suitable and comprehensive for my thesis.

This study analyzed English Major Students' perceptions of the Chat GPT IELTS Writing Correction Tool in their writing process, gathering data through an online survey and interviews.

Questionnaires are widely utilized in scientific research as effective tools for collecting data on opinions, perceptions, and behaviors through a structured series of questions They offer an efficient and convenient method for gathering substantial amounts of relevant information, making them essential for various studies.

In a study by Shima Balouchi (2020), key factors affecting learners' intentions to use technology for informal English learning were examined, highlighting student satisfaction with augmented reality applications The research employed an online survey using a five-point Likert scale to gauge responses effectively.

(1932) through the Google Forms website The five points of the Likert scale applied in the survey are:

The survey, comprising 30 multiple-choice questions organized into four sections, aimed to address specific study questions, resulting in 46 responses from a total population of 54 Utilizing the Likert Scale, a psychological assessment tool, allowed for the measurement of emotions, attitudes, and opinions through a series of related statements, where respondents indicated their level of agreement based on personal experiences The choice of the Likert Scale facilitated straightforward data entry into the Statistical Package for Social Science (SPSS), avoiding complex procedures Additionally, the 1 to 5 rating scale provided enhanced accuracy in capturing participants' viewpoints compared to a simple Yes-No format.

To enhance the depth of user opinions, this study included semi-structured interviews with five English-major students who utilized the Chat GPT IELTS Writing Correction Tool during their preparation for the IELTS exam The interview comprised five open-ended questions, allowing participants to share their personal experiences without any predetermined standards This format encouraged detailed responses rather than simple Yes-No answers, enabling the research to capture the participants' perspectives and gather high-quality insights.

The main data was collected through the model described below:

Figure 3 The process used in this study

A survey was conducted from March 19, 2024, involving 46 English-Major students at the Banking Academy, following consultations with supervising teachers and relevant studies To collect feedback, survey forms were distributed through platforms like Facebook, Zalo, and Messenger.

Figure 4 The four sections of developing the research’s survey

1 • Conducting the online questionnaire survey data

• Scheduling online meetings with participants for the semi-structured interview about the Chat GPT IELTS Writing Correction Tool

3 • Conducting online meetings with participants

4 • Collecting the online questionnaire survey data

5 •Collecting the semi- structured interview data

For qualitative data, the study collected information through semi-structured interviews with 5 English-Major Students at the Banking Academy However, only

4 students participated in the interviews, and 1 student declined to be interviewed due to personal reasons.

Data analysis

Prior to analyzing survey data through various chart formats, several validation measures were implemented to ensure the reliability and accuracy of the survey results The survey table was first reviewed by supervising teachers to confirm the validity of the data and the relevance of the questions to user experiences Additionally, the data underwent validation using Cronbach’s Alpha coefficient, which ranges from 0 to 1; a coefficient of 0.7 or higher indicates high reliability, while values of 0.95 or above necessitate the exclusion of those variables from the survey Finally, the collected data was analyzed using SPSS version 25, a widely used tool in research for examining numerical data and identifying relationships between variables.

First Section: Automated Essay Scoring (AES)

Second Section: The Chat GPT IELTS Writing Correction

Assessing the Chat GPT IELTS Writing Correction Tool according to TAM: a Usefulnes b Ease of Use c Expectation Confirmation

Fourth Section: English-Major Students' expectations towards the use of the Chat GPT IELTS Writing Correction Tool

The analysis of interview data employed a deductive approach alongside thematic analysis to validate initial assumptions, resulting in precise and reliable research findings.

FINDINGS AND DISCUSSION

4.1.1 Findings and Discussions of AES

Table 4.1 Levels of agreement or disagreement

Table 4.2 Reliability Scale of AES

Cronbach's Alpha if Item Deleted

Scoring (AES) can be an effective tool for IELTS

A2 Immediate feedback is a crucial aspect of AES

Writing Skill process, using AES saves money

Writing Skill process, using AES saves time

Writing Skill process, AES can create for learners an interactive environment

AES is totally new to me

The table 4.2 illustrates correlation coefficients of variables within the factor

English-major students perceive the Automated Essay Scoring system as reliable, supported by a Cronbach's Alpha coefficient of 0.824, which exceeds the acceptable threshold of 0.7 Additionally, all Corrected Item-Total Correlation values are above 0.3, further confirming the system's reliability.

Table 4.3 English-Major Students’ perceptions towards using the Automated Essay Scoring

1 Automated Essay Scoring (AES) can be an effective tool for

2 Immediate feedback is a crucial aspect of AES 4.15 0.51 2

3 In learning the IELTS Writing Skill process, using AES saves money

4 In learning the IELTS Writing Skill process, using AES saves time

5 In learning the IELTS Writing Skill process, AES can create for learners an interactive environment

6 Improving the IELTS Writing Skill by using AES is totally new to me

Table 4.3 illustrates information about English-Major Students’ perceptionstowards using Automated Essay Scoring (AES) In table 4.3, it can be seen that the statement

Automated Essay Scoring (AES) proves to be a valuable resource for enhancing IELTS Writing skills, achieving the highest mean score of 4.24 and a standard deviation of 0.43 Immediate feedback, a vital feature of AES, follows closely with a mean value of 4.15 and a standard deviation of 0.51 Conversely, the perception of AES as a completely new approach to improving IELTS Writing skills received the lowest ranking, with a mean of 2.8 and a standard deviation of 0.96.

"In learning the IELTS Writing Skill process, AES can create for learners an interactive environment." with a mean value of 3.63 and SD of 0.77

The data reveals that English-major students have a favorable perception of AES, with an average rating of 3.75, indicating a high level of acceptance among the students.

English-Major students exhibit a high level of acceptance towards the effectiveness of Automated Essay Scoring (AES), primarily due to the immediate feedback it provides on their essays, which aids in improving their IELTS writing skills Their positive experiences with AES foster optimism in the learning process, leading to greater acceptance of this technology However, many students do not perceive AES as entirely novel, as the rapid advancement of information technology has granted widespread access to online academic resources With nearly every student owning a laptop, engaging with artificial intelligence tools to enhance their learning and IELTS writing skills has become increasingly accessible.

4.1.2 Findings and Discussions of Chat GPT IELTS Writing Correction Tool

Table 4.4 Reliability Scale of Chat GPT IELTS Writing Correction Tool

Scale Mean if Item Deleted

Scale Variance if Item Deleted

B1 I use the Chat GPT IELTS

Writing Correction Tool once to several times a day

B2 I use the Chat GPT IELTS

Writing Correction Tool once to three times a week

B3 I spend from 15 to 20 minutes using the Chat GPT IELTS

B4 I spend more than 20 minutes using the Chat GPT IELTS

B5 I like all the features in the

Writing Task 1 and 2, Analysis of prompts and essays, Identify errors, analysis of mistakes, grading and giving comments according to IELTS Writing Band

B6 I think the feedback in the

Correction Tool is perfect (the tool The will evaluate each criterion: Lexical & Grammatical errors/Task Response/ Coherence

Grammatical range & accuracy and show the ban

A study on English-major students' perceptions of Automated Essay Scoring revealed a reliable Likert Scale, evidenced by a Cronbach's Alpha coefficient of 0.93, surpassing the acceptable threshold of 0.7 Additionally, all Corrected Item-Total Correlation values exceeded the minimum requirement of 0.3, indicating strong internal consistency.

Table 4.5 English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool in general

7 I use the Chat GPT IELTS Writing Correction Tool once to several times a day

8 I use the Chat GPT IELTS Writing Correction Tool once to three times a week

9 I spend from 15 to 20 minutes using the Chat GPT IELTS

10 I spend more than 20 minutes using the Chat GPT IELTS

11 I like all the features in the Chat GPT IELTS Writing

Correction Tool (Outline for Writing Task 1 and 2, Analysis of prompts and essays, Identify errors, analysis of mistakes, grading and giving comments according to IELTS Writing

12 I think the feedback in the Chat GPT IELTS Writing

Correction Tool is perfect (the tool The will evaluate each criterion: Lexical & Grammatical errors/Task Response/

Coherence & Cohesion/ Lexical resource/ Grammatical range

& accuracy and show the band score alongside each criterion with detailed feedback)

The table 4.5 depicts the information about English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool in general The highest rank

The Chat GPT IELTS Writing Correction Tool has received positive feedback from English-Major Students, with an average satisfaction score of 3.76 Key features such as task evaluation across criteria like Task Achievement, Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy garnered a mean score of 4.43 Additionally, participants appreciated the tool's functionalities, including outlines for Writing Tasks 1 and 2, prompt analysis, error identification, and grading based on IELTS Writing Band Descriptors, which received a mean score of 4.02 Overall, students express high satisfaction with the tool as a valuable resource for improving their IELTS writing skills.

In conclusion, English majors express significant satisfaction with the scores and assessments provided by the Chat GPT IELTS Writing Correction Tool Utilizing IELTS Writing Band Descriptors, this tool delivers accurate and detailed feedback, enabling students to identify areas for improvement Additionally, it offers comprehensive guidance on essential learning steps, such as prompt analysis and outlining for both Task 1 and Task 2.

2, which makes students express satisfaction with this tool

4.1.3 Findings and Discussions of Perceived of Ease of Use

Table 4.6 Reliability Scale of Perceived of Ease of Use

C2 The using instructions in the Chat GPT IELTS Writing

Tool simple in design and easy to navigate

C4 I can easily switch between the grading sections of Task 1 and Task 2 and quickly review the grading history

IELTS Writing Skill using the Chat GPT IELTS Writing

English-major students view Automated Essay Scoring as a reliable tool, evidenced by a Cronbach's Alpha coefficient of 0.93, surpassing the acceptable threshold of 0.7 Additionally, all Corrected Item-Total Correlation values exceed 0.3, further supporting its reliability.

Table 4.7 English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool as an easy learning tool

13 Registering to the Chat GPT IELTS Writing Correction Tool was convenient

14 The using instructions in the Chat GPT IELTS Writing

15 I find the Chat GPT IELTS Writing Correction Tool simple in design and easy to navigate

16 I can easily switch between the grading sections of Task 1 and

Task 2 and quickly review the grading history

17 I enjoyed learning the IELTS Writing Skill using the Chat

GPT IELTS Writing Correction Tool

Table 4.7 reveals English-Major Students' positive perceptions of the Chat GPT IELTS Writing Correction Tool as a user-friendly learning resource The tool received the highest rating for its simple design and ease of navigation, with a mean score of 4.57 Additionally, the convenience of registering for the tool was noted, earning a score of 4.45 Students also appreciated the ability to easily switch between grading sections for Task 1 and Task 2, as well as the enjoyable learning experience it provided Overall, the mean score of 4.47 indicates a strong acceptance of the Chat GPT IELTS Writing Correction Tool as an effective aid for practicing IELTS writing skills.

Students can easily access the Chat GPT IELTS Writing Correction Tool by registering with their Gmail, Facebook, or phone number, receiving 5 free trials upon account creation The tool allows users to enter their essays or upload prompts to receive helpful outlines With straightforward usage instructions and a user-friendly design that clearly differentiates between Task 1 and Task 2, this tool enhances the learning experience, making it a preferred choice for students seeking assistance with their writing.

4.1.4 Findings and Discussions of Perceived of Usefulness

Table 4.8 Reliability Scale of Perceived of Usefulness

Writing Correction Tool improved my overall task response

Writing Correction Tool increased my ability to organize and structure an essay

Writing Correction Tool has given a chance to expand vocabulary range

Writing Correction Tool improved my grammatical accuracy

Writing Correction Tool motivated me to control word count

Writing Correction Tool has given me the courage to demonstrate critical thinking

Correction Tool is a helpful tool

English-major students' perceptions of Automated Essay Scoring demonstrate high reliability, as indicated by a Cronbach's Alpha coefficient of 0.97, significantly surpassing the acceptable threshold of 0.7 Additionally, all Corrected Item-Total Correlation values exceed the minimum requirement of 0.3, further supporting the validity of the Likert Scale used in the study.

Table 4.9 English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool as a useful learning tool

18 The Chat GPT IELTS Writing Correction Tool improved my overall task response

19 The Chat GPT IELTS Writing Correction Tool increased my ability to organize and structure an essay

20 The Chat GPT IELTS Writing Correction Tool has given a chance to expand vocabulary range

21 The Chat GPT IELTS Writing Correction Tool improved my grammatical accuracy

22 The Chat GPT IELTS Writing Correction Tool motivated me to control word count

23 The Chat GPT IELTS Writing Correction Tool has given me the courage to demonstrate critical thinking

24 I believe that the Chat GPT IELTS Writing Correction Tool is a helpful tool

Table 4.9 presents English-major students' perceptions of the Chat GPT IELTS Writing Correction Tool as an effective learning resource The highest-rated statement, "I believe that the Chat GPT IELTS Writing Correction Tool is a helpful tool," received a mean score of 4.33 with a standard deviation of 0.89 Following closely, the statement "The Chat GPT IELTS Writing Correction Tool improved my grammatical accuracy" achieved a mean of 4.13 and a standard deviation of 0.91 Conversely, the statement "The Chat GPT IELTS Writing Correction Tool has given me the courage to demonstrate critical thinking" received the lowest rating at 3.96 Overall, students' average acceptance of the tool's usefulness is 4.09, indicating a relatively high level of approval.

The Chat GPT IELTS Writing Correction Tool is often ranked low for fostering critical thinking due to its primary focus on vocabulary, grammar, and essay structure While it aids in writing improvement, developing critical thinking skills is better achieved through extensive reading, diverse learning sources, and face-to-face conversations.

4.1.5 Findings and Discussions of Expectation Confirmation

Table 4.10 Reliability Scale of Expectation Confirmation

Scale Mean if Item Deleted

Writing Correction Tool was as I expected when I registered it

E2 I want to be graded more by the Chat GPT IELTS

Writing Correction Tool, so I buy more essay scoring credits on the website

E3 I have a plan of taking the IELTS exam after my

IELTS Writing grade by the tool has consistently reached my expected band

Correction Tool to practice my IELTS Writing tasks in the future

Writing Correction Tool has increased my interest in practicing IELTS Writing

Writing Correction Tool has encouraged me to use other

English-major students' perceptions of Automated Essay Scoring were assessed using a reliable Likert Scale, evidenced by a Cronbach's Alpha coefficient of 0.97, significantly exceeding the acceptable threshold of 0.7 Additionally, all Corrected Item-Total Correlation values surpassed the minimum requirement of 0.3, further confirming the scale's reliability.

Table 4.11 English-Major Students’ expectations regarding the Chat GPT IELTS Writing Correction Tool

25 The Chat GPT IELTS Writing Correction Tool was as I expected when I registered it

26 I want to be graded more by the Chat GPT IELTS Writing

Correction Tool, so I buy more essay scoring credits on the website

27 I have a plan of taking the IELTS exam after my IELTS

Writing grade by the tool has consistently reached my expected band

28 I consider using the Chat GPT IELTS Writing Correction Tool to practice my IELTS Writing tasks in the future

29 The Chat GPT IELTS Writing Correction Tool has increased my interest in practicing IELTS Writing Skill

30 The Chat GPT IELTS Writing Correction Tool has encouraged me to use other AES systems

Table 4.11 reveals the expectations of English-major students regarding the Chat GPT IELTS Writing Correction Tool The tool significantly boosts interest in practicing IELTS writing, achieving the highest average score of 4.52 Additionally, students express a desire for more grading by the tool, reflected in a mean score of 4.37, which motivates them to purchase more essay scoring credits Conversely, the statement regarding encouragement to use other Automated Essay Scoring (AES) systems received the lowest ranking, with an average score of 3.13.

In conclusion, learners’ initial expectations for would be surpassed after using theChat GPT IELTS Writing Correction Tool This positive expectation confirmation

CONCLUSION AND RECOMMENDATIONS

This chapter will offer a brief summary of findings and recommendations, which encounters the results of the research and discusses my expectancy of readers, students, teachers, and the future researchers

This study, conducted at the English Language Department of the Banking Academy, utilized an online survey and semi-structured interviews with 46 English language students to explore perceptions of the Chat GPT IELTS Writing Correction Tool The research focused on three key factors: Perceived Usefulness, Perceived Ease of Use, and Expectation Confirmation, aiming to assess how these elements influence the improvement of IELTS writing proficiency.

Research shows that English language students find the Chat GPT IELTS Writing Correction Tool useful, easy to use, and aligned with their expectations The tool effectively aids learners in enhancing their IELTS writing skills, including vocabulary, grammar, and idea development However, it has limitations in fostering logical reasoning and argumentation, as it offers pre-programmed suggestions that do not fully capture the writer's individual style and cannot replace an instructor's role Additionally, users may experience drawbacks such as distraction and eye strain from extended screen time.

The TAM model suggests that, in addition to Perceived Usefulness and Perceived Ease of Use, factors such as affordability and enjoyment of use can significantly impact English language students' acceptance of the Chat GPT IELTS Writing Correction Tool.

The research process faced limitations, primarily due to a two-month timeframe and a small sample of English language students A larger sample size, including all IELTS candidates in Hanoi, would yield more accurate conclusions Consequently, the acceptance of the Chat GPT IELTS Writing Correction Tool as a useful and user-friendly resource may not reflect the opinions of all IELTS learners who have utilized it in their studies.

One limitation of the study is the reliance on a single data collection method, specifically an online survey questionnaire and semi-structured interviews While additional interviews aimed to validate students' opinions, they may not fully capture the students' acceptance of the Chat GPT IELTS Writing Correction Tool To enhance the reliability of the findings, I compared the results with previous related studies.

This research aims to provide IELTS learners with effective learning methods and valuable support tools, transforming the learning experience from a burdensome task into an engaging process AI tools like the Chat GPT IELTS Writing Correction Tool can significantly enhance the learning journey, especially since obtaining an IELTS certificate is crucial for English language students pursuing dreams such as studying abroad, securing scholarships, or conducting research The Chat GPT IELTS Writing Correction Tool stands out for its ability to save time and reduce anxiety regarding writing skills, making it an essential resource for learners at the Banking Academy and beyond.

A study on the perceptions of English-major students at the Banking Academy regarding the Chat GPT IELTS Writing Correction Tool aims to provide valuable insights for educators The research highlights the challenges and benefits students experience while using this tool, which can inform teachers on its integration into their teaching strategies By understanding student psychology, educators can optimize IELTS writing practice and create engaging learning activities that foster enthusiasm for using the tool This approach encourages students to actively enhance their writing skills, ultimately improving their IELTS performance.

This study aims to establish a foundational framework for future research on the awareness and strategies for utilizing Automated Essay Scoring (AES) tools in English language learning and certification preparation The findings provide valuable insights for researchers to guide their investigations into the integration of AI tools in both language learning and teaching methodologies, potentially leading to the development of innovative and effective teaching approaches for educators and students alike.

REFERENCES Al-Emran, M., Mezhuyev, V., & Kamaludin, A (2018) Technology Acceptance Model in

M-learning context: A systematic review Computers and Education/Computers & Education, 125, 389–412 https://doi.org/10.1016/j.compedu.2018.06.008

This study by Al-Azawei, Parslow, and Lundqvist (2016) examines how learning styles influence perceptions of a blended e-learning system (BELS) using an adapted technology acceptance model (TAM) Implemented at an Iraqi university, this research highlights the integration of Moodle with traditional face-to-face classrooms to enhance learning flexibility The study involved 210 undergraduate students and utilized the Index of Learning Styles Questionnaire to assess the relationship between learner satisfaction and technology adoption The findings suggest that while the model effectively predicts intention to use and perceived satisfaction, psychological differences among learners did not significantly impact their satisfaction or e-learning adoption The research emphasizes the importance of accommodating individual learning styles in educational hypermedia systems to improve user experience and satisfaction.

Marzuki, Widiati, U., Rusdin, D., Darwin, & Indrawati, I (2023) The impact of AI writing tools on the content and organization of students’ writing: EFL teachers’ perspective Cogent Education, 10(2) https://doi.org/10.1080/2331186x.2023.2236469

Human verification (n.d.) https://www.semanticscholar.org/paper/Artificial-intelligence-

The role of CHATGPT as AI English Writing Assistant: A study of student’s perceptions, experiences, and satisfaction (2024) Annals of Human and Social Sciences, 5(I) https://doi.org/10.35484/ahss.2024(5-i)39

Kurniati, E Y., & Fithriani, R (2022) Post-Graduate students’ perceptions of Quillbot utilization in English Academic Writing class JELTL (Journal of English

Language Teaching and Linguistics), 7(3), 437 https://doi.org/10.21462/jeltl.v7i3.852

Human verification (n.d.-b) https://www.semanticscholar.org/paper/Do-Digital-Writing-

Tools-Deliver-Student-of-Writing-Nobles-

Baskara, F R (2023) Integrating ChatGPT into EFL writing instruction: Benefits and challenges International Journal of Education and Learning, 5(1), 44 – 55 https://doi.org/10.31763/ijele.v5i1.858

Human verification (n.d.-c) https://www.semanticscholar.org/paper/The-role-of-

ChatGPT-in-scientific-communication%3A-Huang-

Tan/b575d7eb2d9edc9765db38d0149d1336f3596ebb

Zhao, X (2022) Leveraging Artificial intelligence (AI) technology for English Writing:

Introducing Wordtune as a Digital writing assistant for EFL writers https://www.semanticscholar.org/paper/Leveraging-Artificial-Intelligence-(AI)- Technology-Zhao/d7aeef9619a774236dbc901459bdc614daee0ed5

Wu, X (2022) Dynamic evaluation of college English writing ability based on AI technology Journal of Intelligent Systems, 31(1), 298–309 https://doi.org/10.1515/jisys-2022-0020

In their 2022 study, Sumakul, Hamied, and Sukyadi explore students' perceptions of artificial intelligence (AI) in writing classes, highlighting its impact on learning and engagement The research, published in Advances in Social Science, Education and Humanities Research, reveals that students view AI as a valuable tool that enhances their writing skills and provides personalized feedback The findings suggest that integrating AI into educational settings can foster a more interactive and supportive learning environment, ultimately benefiting students' academic performance For more insights, refer to the full article at https://doi.org/10.2991/assehr.k.220201.009.

Nazari, N., Shabbir, M S., & Setiawan, R (2021) Application of Artificial Intelligence powered digital writing assistant in higher education: randomized controlled trial Heliyon, 7(5), e07014 https://doi.org/10.1016/j.heliyon.2021.e07014

Looi, C., Seow, P S K., Zhang, B., So, H., Chen, W., & Wong, L (2010) Leveraging mobile technology for sustainable seamless learning: a research agenda British Journal of Educational Technology, 41(2), 154–169 https://doi.org/10.1111/j.1467-

Ahmed, H B E (2016) Duolingo as a Bilingual Learning App: A case study Social

Science Research Network https://doi.org/10.2139/ssrn.2814822

My name is Dang Thu Thao, a K23 student at the Faculty of Foreign Language, Banking Academy of Vietnam I am conducting my graduation paper on "English-Major Students’ Perceptions Towards Using the Chat GPT IELTS Writing Correction Tool for Improving IELTS Writing Skills." I invite you to participate in this survey, which aims to collect data for my project by completing the questionnaire Please answer the questions honestly, following the provided instructions.

Your answer will be kept strictly confidential and anonymous You might withdraw at any time during this questionnaire

If you have any questions, please contact me

Thank you in advance for your participation!

Have you ever used the Chat GPT IELTS Writing Correction Tool for your learning? o Yes o No

For each statement below, please select from 1 to 5 to indicate how much you disagree or agree

Table 3.1: Construction of the questionnaire

Questions 1 2 3 4 5 English-Major Students’ perceptions towards using the Automated Essay Scoring

1 Automated Essay Scoring (AES) can be an effective tool for IELTS

2 Immediate feedback is a crucial aspect of AES

English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool in general

9 I spend from 15 to 20 minutes using the Chat GPT IELTS Writing

10 I spend more than 20 minutes using the Chat GPT IELTS Writing

11 I like all the features in the Chat GPT IELTS Writing Correction

Tool (Outline for Writing Task 1 and 2, Analysis of prompts and essays, Identify errors, analysis of mistakes, grading and giving comments according to IELTS Writing Band Descriptors)

12 I think the feedback in Chat GPT IELTS Writing Correction Tool is perfect (the tool The will evaluate each criterion: Lexical &

Grammatical errors/Task Response/ Coherence & Cohesion/

Lexical resource/ Grammatical range & accuracy and show the band score alongside each criterion with detailed feedback)

English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool as an easy learning tool

14 The using instructions in the Chat GPT IELTS Writing Correction

16 I can easily switch between the grading sections of Task 1 and Task

2 and quickly review the grading history

17 I enjoyed learning the IELTS Writing Skill using the Chat GPT

English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool as a useful learning tool

English-Major Students’ expectations regarding the Chat GPT IELTS Writing Correction Tool

27 I have a plan of taking the IELTS exam after my IELTS Writing grade by the tool has consistently reached my expected band

29 The Chat GPT IELTS Writing Correction Tool has increased my interest in practicing IELTS Writing Skill

30 The Chat GPT IELTS Writing Correction Tool has encouraged me to use other AES systems

Table 3.2: Data from the questionnaire coding frame

English-Major Students’ perceptions towards using the Automated Essay Scoring

1 Automated Essay Scoring (AES) can be an effective tool for

2 Immediate feedback is a crucial aspect of AES A2

English-Major Students’ perceptions towards the Chat GPT IELTS Writing Correction Tool in general

9 I spend from 15 to 20 minutes using the Chat GPT IELTS

10 I spend more than 20 minutes using the Chat GPT IELTS

11 I like all the features in the Chat GPT IELTS Writing Correction

Tool (Outline for Writing Task 1 and 2, Analysis of prompts and essays, Identify errors, analysis of mistakes, grading and giving comments according to IELTS Writing Band

12 I think the feedback in Chat GPT IELTS Writing Correction

Tool is perfect (the tool The will evaluate each criterion:

Lexical & Grammatical errors/Task Response/ Coherence &

Cohesion/ Lexical resource/ Grammatical range & accuracy and show the band score alongside each criterion with detailed feedback)