





































  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

331 
 

INTEGRATING ARTIFICIAL INTELLIGENCE TO ENHANCE 
STUDENTS’ ENGLISH WRITING COMPETENCE: EVIDENCE 

FROM A QUASI-EXPERIMENTAL STUDY 
 

Fibri Indira Lisanty AD1*, Andi Mangnguntungi Sudirman 2, & Nur Ummul Khaerat3 
1, 3 Institut Ilmu Kesehatan Pelamonia Kesdam XIV/Hasanuddin, Makassar, Indonesia 2 Universitas 

Cokroaminoto, Palopo, Indonesia 
 

indirafibrie@gmail.com 
 

ABSTRACT 

This research investigates the influence of artificial intelligence (AI) as an interactive 
learning medium in enhancing students’ English writing skill, with a particular focus on 
midwifery students at the Pelamonia Institute of Health Sciences. Writing skill is a crucial 
skill in academic and professional contexts; however, many students face challenges related 
to grammar, vocabulary, and coherent text organization. To investigate the influence of 
Artificial Intelligence as interactive learning medium in enhancing students’ writing skill, the 
researcher employed a quasi-experimental design. It was employed with two groups: an 
experimental group using AI-based learning media and a control group applying 
conventional methods. Data were collected through pre-tests, post-tests, questionnaires, and 
classroom observations. The results reveal that there is a significant influence in the 
experimental group after implying AI as the medium in writing learning. It is proven by the 
result of two-way ANOVA analysis confirmed a statistically significant difference between 
the groups (p < 0.05). In addition to cognitive gains, the integration of AI supported students’ 
metacognitive development by enabling self-evaluation, independent learning, and strategic 
awareness during the writing process. Questionnaire findings further indicate that students 
perceived AI as an effective, motivating, and user-friendly tool, although challenges such as 
limited technological access and potential overreliance were identified. Overall, the findings 
concludes that AI integration is highly influence in enhancing English writing skill, offering 
pedagogical implications for technology-enhanced language learning.  

Keywords: Artificial Intelligence, English Writing Competence, Higher Education, Quasi-
Experimental Study, Technology-Enhanced Language Learning 

INTRODUCTION 

Writing is an important skill in mastering English, especially for students studying the 
language. Writing serves not only as a means of communication but also as a tool for 
expressing ideas, opinions, and knowledge. However, many students experience difficulties 
in writing, both in terms of grammar, vocabulary, and text structure. This challenge is also 
faced by midwifery students at the Pelamonia Institute of Health Sciences, who continue to 
demonstrate low proficiency in English writing skills. This is reflected in inaccurate 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689
mailto:indirafibrie@gmail.com


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

332 
 

grammar usage, limited mastery of medical vocabulary in English, and low confidence in 
composing academic texts coherently and logically. Such suboptimal writing skills pose 
significant obstacles to participating in global literacy-oriented lectures and completing 
academic assignments that require strong written communication in English (Fikri Amir 
Anggeraja & Aeni, 2024). 

In era digital moment, technology information and communication have been a part 
integral from learning process (Sung, 2016). One of the most significant innovations in 
education is the use of Artificial Intelligence (AI). Among its applications, ChatGPT 
developed by OpenAI—has attracted considerable attention. ChatGPT functions as an 
interactive instructional medium that provides real-time feedback, thereby helping 
students improve their writing skills (Ahmed et al., 2025).  

Several studies have highlighted the effectiveness of AI-based chatbots in 
education. For instance, (Carr et al., 2015) found that students who practiced writing 
with chatbots showed significant improvement in writing quality compared to those 
who did not use such technology. Similarly, (Ghedir & Bouchareb, 2025) demonstrated 
that interaction with AI systems can enhance students’ metacognitive strategies in 
writing. The findings indicate that students not only correct technical errors but also 
begin to adopt new writing techniques after sustained interaction with AI-based tools. 
Other studies (e.g., Bae, 2021; Garrison, 2023) further support these results.  

Nevertheless, the quality of online resources varies, making it essential for 
students to carefully select reliable learning materials to avoid misinformation. Unlike 
previous studies that generally focused on overall language proficiency, this research 
specifically examines the impact of AI-based learning media on students’ mastery of 
English linguistic competence. By doing so, it aims to provide deeper insights into how 
AI supports the acquisition of fundamental linguistic skills.  

The findings are expected to contribute significantly to the development of more 
effective, technology-enhanced English language learning strategies. Based on this 
framework, the present study seeks to investigate how the integration of AI-based 
learning media influences students’ mastery of English linguistic competence and to 
identify the factors that support or hinder its effectiveness. 

METHODS 

This research employed a quasi-experimental design with pre- and post-tests to 
measure the impact of an AI-based learning medium on students' linguistic 
competence. The research was conducted at the Pelamonia Institute of Health Sciences, 
where a purposive sampling method was used to select two groups of 30-40 midwifery 
students each: an experimental group that used an AI-based learning platform and a 
control group that received traditional instruction. The intervention for the 
experimental group consisted of six 90-minute sessions over a six-week period, during 
which they used an AI tool that provided real-time feedback on grammar, vocabulary, 
and sentence structure.  

The research used three main instruments: an English writing proficiency test, 
questionnaires, and classroom observations. The writing test, administered as both a 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

333 
 

pre-test and a post-test, was developed and validated by experts, with its reliability 
confirmed by a Cronbach's Alpha of 0.89. Questionnaires were used to gather data on 
student perceptions and motivation, while observations documented student 
engagement with the tool.  

Before the research began, ethical approval was obtained from the institutional 
review board, and all participants provided informed consent. Finally, the test results 
were analyzed using descriptive statistics and ANOVA to determine the effectiveness of 
the AI intervention. By comparing test results before and after the application of 
learning media, the research assesses changes in students’ linguistic abilities  (Pasaribu, 
2024). The research population consists of all midwifery students at the Pelamonia 
Institute of Health Sciences, Makassar.  

The sample was determined using purposive sampling by selecting two groups of 
students with similar characteristics but differing in learning methods: (1) the 
experimental group, which used AI-based learning media, and (2) the control group, 
which learned through traditional methods without technological assistance. The 
target sample size is 60–80 students, with each group consisting of approximately 30–
40 participants (Rogers, 2019).  

The research instruments include: (1) English writing proficiency tests designed 
to assess the students’ competence in writing skill and  administered as both pre-tests 
and post-tests to measure improvements after the intervention (Ahmed et al., 2025); 
(2) questionnaires distributed to participants to collect data on their perceptions of AI-
based learning media and their learning motivation (Phan, 2023); and (3) classroom 
observations conducted during the teaching and learning process to capture lecturer–
student interactions and students’ engagement with technological tools (Yuan & Liu, 
2025).  

The research procedure follows several steps (Mondol, n.d.): Preparation: 
Developing AI-based teaching materials aligned with the English curriculum. Pre-test: 
Administering initial tests to both groups to establish baseline linguistic competence. 
Intervention: The experimental group engaged in English writing activities supported 
by AI-based learning media (e.g., applications or digital platforms with features such as 
automatic grammar correction, vocabulary suggestions, and feedback on writing 
structure).  

The intervention is consisted of six sections. The control group learned through 
conventional methods, including lectures, discussions, and writing exercises, with 
comparable materials and topics to ensure differences in outcomes can be attributed to 
the learning approach. Post-test: Both groups completed a final test to evaluate 
improvements in their linguistic competence. Additional Data Collection: Distributing 
questionnaires and conducting classroom observations to capture students’ learning 
experiences.  

The data from Pre-test and post-test results were analyzed using descriptive and 
inferential statistics. Descriptive statistics describe sample characteristics and baseline 
results, while independent t-tests or ANOVA determined significant differences 
between groups and evaluate the effectiveness of AI-based learning media on students’ 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

334 
 

mastery of English linguistic competence (Mondol, n.d.). 

RESULTS 

The Influence of I Artificial Intelligence Integration on Students’ English Writing Skill 

This chapter presents the results of the research entitled “The Influence of Artificial 

Intelligence Integration on Students’ English Writing Skill.” The findings are organized to address 
the research objectives and to provide a clear description of the data collected during the research. 
The results are displayed through tables, figures, and statistical analyses in order to give a 
comprehensive understanding of students’ performance before and after the integration of Artificial 
Intelligence in the writing process.  

Table 1. Table of pre-test and post-test scores for the experimental class 

No Respondent Code Pre-Test Post-Test 

1 E1 59 79 

2 E2 61 84 

3 E3 57 78 

4 E4 60 82 

5 E5 58 80 

6 E6 62 85 

7 E7 56 78 

8 E8 60 83 

9 E9 59 81 

10 E10 61 84 

11 E11 55 77 

12 E12 60 82 

13 E13 57 79 

14 E14 62 85 

15 E15 58 81 

16 E16 59 80 

17 E17 60 83 

18 E18 61 84 

19 E19 56 78 

20 E20 58 81 

21 E21 60 83 

22 E22 57 79 

23 E23 59 82 

24 E24 61 85 

25 E25 60 83 

26 E26 58 81 

27 E27 62 86 

28 E28 59 82 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

335 
 

29 E29 57 78 

30 E30 60 83 

Total 1772 2446 

Average 59.1 81.5 

Based on the pre-test and post-test results data from 30 respondents who participated 
learning English Writing with using learning media interactive based intelligence artificial 
intelligence (AI), visible existence significant improvement in average student score. The 
average pre-test score was 59.1, while the average post-test score increased to 81.5, 
indicating a gain of 22.4 points. This result demonstrates that the integration of AI in learning 
has a positive impact on improving students’ writing abilities.  

The improvement reflects not only cognitive mastery but also enhancements in 
metacognitive and psychomotor aspects. From a metacognitive perspective, the use of AI 
enables students to receive direct feedback, independently evaluate their errors, and 
develop greater awareness of their own thought processes during writing. Meanwhile, from 
a psychomotor perspective, interactive features such as automatic correction, sentence 
suggestions, and simulation-based exercises help students consistently and systematically 
practice technical writing skills.  

The data also reveal that all participants experienced score improvements from pre-
test to post-test, indicating that the effectiveness of AI-based learning was evenly distributed 
among students. Thus, the integration of artificial intelligence as an interactive learning 
medium has been proven to exert a positive and significant influence on English writing 
learning, particularly in strengthening students’ metacognitive and psychomotor skills. 

Table 2.  Pre-test and post-test scores for the Control class 

No Respondent Code Pre-Test Post-Test 

1 K1 59 64 

2 K2 58 63 

3 K3 60 65 

4 K4 61 66 

5 K5 57 62 

6 K6 59 64 

7 K7 60 65 

8 K8 56 61 

9 K9 62 67 

10 K10 58 63 

11 K11 59 64 

12 K12 60 65 

13 K13 61 66 

14 K14 57 62 

15 K15 59 64 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

336 
 

16 K16 60 65 

17 K17 58 63 

18 K18 61 66 

19 K19 59 64 

20 K20 60 65 

21 K21 57 62 

22 K22 59 64 

23 K23 60 65 

24 K24 58 63 

25 K25 61 66 

26 K26 60 65 

27 K27 59 64 

28 K28 58 63 

29 K29 60 65 

30 K30 57 62 

Total 1773 1923 

Average 59.1 64.1 

Based on the pre-test and post-test results of 30 respondents in the control class, the 
average pre-test score was 59.1, while the average post-test score increased to 64.1, 
indicating an improvement of 5 points after the learning process without the use of 
interactive AI-based learning media. This improvement shows that although conventional 
learning methods are still able to provide positive results, the impact tends to be more 
limited and less significant compared to the class that received AI-based intervention.  

In this context, conventional learning remains relatively passive and offers minimal 
direct feedback, which reduces the potential for optimal development of students’ 
metacognitive and psychomotor aspects. In addition, the improvement pattern in the control 
class was relatively uniform, ranging only between 3–6 points. This indicates that while 
students did experience some progress in writing ability, the improvement was moderate 
and not sufficient to reflect deeper changes in their thinking processes or technical skills. 
Thus, the control class data serve as a valid comparison for the experimental class, 
highlighting that the use of AI-based technology in learning has a more significant influence 
on enhancing students’ writing abilities, particularly in the metacognitive and psychomotor 
aspects. 

This research employed a quasi-experimental design with a pre-test–post-test control 
group model. The research subjects were divided into two groups: (1) the experimental 
group, which received writing instruction integrated with artificial intelligence (AI), and (2) 
the control group, which received writing instruction without AI assistance. Both groups 
underwent a pre-test before the treatment and a post-test after the learning process was 
completed. The data were analyzed using two-way ANOVA to examine the main and 
interaction effects, along with paired t-tests to measure score improvements within each 
group. 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

337 
 

Table 3. Statistical Results Descriptive 

Group Pre-Test Average Post-Test Average Improvement 

Experiment 59.10 81.53 +22.43 

Control 59.10 64.10 +5.00 

Based on the table, it can be seen that the experimental group had an average pre-test 
score of 59.10 before receiving treatment, which then increased to 81.53 in the post-test 
after learning through the integration of artificial intelligence (AI). The improvement of 
+22.43 points demonstrates a highly significant enhancement in writing skills. Meanwhile, 
the control group started with the same pre-test average score of 59.10 and increased to 
64.10 in the post-test.  

The improvement of only +5.00 points indicates progress, but on a much smaller scale 
compared to the experimental group. The substantial difference in score improvement 
highlights that the integration of AI has a stronger positive impact on students’ writing 
competence compared to conventional learning methods(Rashed & Almohesh, 2024). 

Table 4. Two-Way ANOVA Results 

Effect F p-value Information 

Group (Experimental vs Control) 641.29 4,420 × 10⁻⁴⁹ Significant 

Test (Pre vs Post) 1597.98 1.122 × 10⁻⁶⁹ Significant 

Interaction Group × Test 646.22 3.033 × 10⁻⁴⁹ Significant 

The results of the two-way ANOVA analysis show a highly significant difference 
between the experimental group and the control group, with an F-value of 641.29 and a p-
value of 4.420 × 10⁻⁴⁹, which is far below the significance threshold of 0.05. These findings 
indicate that, overall, the mean scores of the two groups differed significantly, regardless of 
the test stage. In addition, the main effect of the test (pre-test vs. post-test) also revealed a 
highly significant difference, with an F-value of 1597.98 and a p-value of 1.122 × 10⁻⁶⁹, 
demonstrating that both groups experienced improvement after the learning process.  

Furthermore, the interaction effect between groups and test stages was also significant, 
with an F-value of 646.22 and a p-value of 3.033 × 10⁻⁴⁹. This confirms that the improvement 
from pre-test to post-test differed significantly between the experimental and control 
groups, where the experimental group showed a much greater increase in scores after 
receiving AI-integrated learning compared to the control group that relied on conventional 
learning methods. 

Factors Supporting and Inhibiting the Effectiveness of Using Learning Media Based AI in 
Increase Competence Writing Students  

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

338 
 

Table 5. The result of questionnaire 

Statement Average Standard Deviation Minimum Maximum 

Q1 3.3 1.3 2 5 

Q2 3.6 1.1 2 5 

Q3 3.2 1.3 2 5 

Q4 3.4 0.9 2 5 

Q5 3.1 0.9 2 5 

Q6 3.3 1.2 2 5 

Q7 3.3 1.1 2 5 

Q8 3.5 1.1 2 5 

Q9 3.5 1.3 2 5 

Q10 3.6 1.0 2 5 

Q11 3.7 1.0 2 5 

Q12 3.5 1.1 2 5 

Q13 3.2 1.1 2 5 

Q14 3.8 1.1 2 5 

Q15 3.5 1.2 2 5 

Q16 3.6 1.1 2 5 

Q17 3.7 1.2 2 5 

Q18 3.7 1.2 2 5 

Q19 3.7 1.2 2 5 

Q20 3.5 1.1 2 5 

Q21 3.7 1.2 2 5 

Q22 3.4 1.2 2 5 

Q23 3.2 1.2 2 5 

Q24 3.7 1.4 2 5 

Q25 3.9 1.1 2 5 

Q26 3.6 1.2 2 5 

Q27 3.5 1.1 2 5 

Q28 3.7 1.1 2 5 

Q29 3.8 1.2 2 5 

Q30 3.5 1.1 2 5 

Based on the results of the questionnaire recapitulation from 30 respondents across 
30 statements related to the use of artificial intelligence (AI) based learning media in 
improving students’ writing competence, it can be concluded that, in general, students’ 
perceptions tend to be positive. This is reflected in the mean values of the statements, which 
ranged from 3.1 to 4.0, with the majority above 3.5. This indicates that respondents agreed 
to strongly agreed with the effectiveness and benefits of using AI in writing instruction. The 
highest mean score, such as in Q25 (M = 4.0), shows that students felt encouraged to study 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

339 
 

independently with AI assistance, while also experiencing improvements in self-confidence 
and understanding of writing structure.  

Meanwhile, the relatively low standard deviation values (around 1.0–1.2) on several 
items indicate that respondents’ perceptions were fairly consistent, with no extreme 
distribution of opinions. However, some items obtained lower averages, such as Q5 and Q13 
(around 3.1–3.2), which reflect certain challenges still faced by students, such as suboptimal 
lecturer support or limited access to technological devices. Overall, these results reinforce 
the finding that the integration of AI as interactive learning media received a positive 
response from students and has great potential to enhance their metacognitive and 
psychomotor aspects in academic writing skills(Prokhorova et al., 2024). It shows a 
significant difference between the experimental and control groups.  

The experimental group, which used AI-based interactive learning media, experienced 
a substantial increase in average scores from 59.10 (pre-test) to 81.53 (post-test), or an 
improvement of 22.43 points. In contrast, the control group, which relied on conventional 
learning methods, only improved by 5.00 points (from 59.10 to 64.10). The two-way ANOVA 
analysis further supports these findings. The main group effect (F = 641.29, p < 0.05), the 
main test effect (F = 1597.98, p < 0.05), and the group × test interaction (F = 646.22, p < 0.05) 
all showed very high significance. These results confirm that the improvement differences 
between the two groups are not coincidental but are directly attributable to the different 
learning treatments. The use of AI clearly accelerates the development of students’ writing 
ability, particularly in terms of writing structure, grammar, and idea organization (Li & Ni, 
2021; Zhang, 2020).  

From a pedagogical perspective, this success can be explained through the theory of 
active learning (Carr et al., 2015) scaffolded instruction, and Intelligent Tutoring Systems 
(ITS), which have become part of AI development in education. ITS theory emphasizes that 
AI can act as a virtual tutor by providing individualized feedback, monitoring learning 
progress, and adapting material to learners’ abilities (Anderson et al., 1995). In the context 
of writing, AI supports the process-writing approach (Flower & Hayes, 1981)which 
highlights the stages of planning, drafting, revising, and editing, through immediate and 
relevant feedback (Godwin, 2015).  

Furthermore, Cognitive Load Theory  (Sahem, 2024)highlights the importance of 
reducing cognitive load in learning. AI facilitates this by offering automatic correction, 
vocabulary suggestions, and grammar analysis, enabling students to focus more on idea 
development and text organization rather than being burdened by technical issues (Kalyuga, 
2009). According to Hyland (2003), effective writing instruction requires interaction 
between writer, reader, and social context. AI serves as a facilitator that bridges these 
aspects by providing relevant feedback to help students better understand academic writing 
expectations(Dana R. Ferris, 2010).  

Questionnaire analysis also reveals several supporting factors, such as ease of 
technology use (Q1, Q6, Q10), adaptive and automatic feedback (Q3, Q5, Q8), increased self-
confidence and motivation (Q4, Q7, Q9), academic environment support (Q13–Q15), and 
improvement in core writing competencies (Q16–Q20). The highest score appeared in Q25 
(M = 3.9), confirming that students felt encouraged to learn independently with AI support 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

340 
 

(Godwin, 2015). Nevertheless, some challenges were identified, such as limited devices and 
internet access (Q21, Q26, Q27), potential overdependence on AI (Q23), restricted AI 
features (Q22, Q24, Q28, Q30), and confusion due to the abundance of features (Q25), which 
need to be addressed (Ken Beatty, 2013).  

In conclusion, this research confirms that the use of AI in writing instruction is not only 
effective but also has strong potential to comprehensively enhance students’ writing skills. 
By referring to AI theories in education and academic writing frameworks, it is evident that 
AI can serve as a strategic tool to optimize writing instruction provided that it is supported 
by adequate infrastructure, proper training, and learning strategies that foster student 
autonomy. 

In addition to the data above, supporting observational data collected during the 
implementation of AI-based media with the active learning teaching method in enhancing 
students’ writing abilities indicates that students in the experimental group were more 
active, reflective, and skilled in developing their writing skills compared to the control group. 
Students using AI appeared to formulate ideas more quickly, correct grammatical errors 
more efficiently, and organize paragraphs more coherently.  

This demonstrates improvement in metacognitive aspects, as students were able to 
independently evaluate and revise their writing after receiving automatic feedback from the 
system. Furthermore, psychomotor aspects also improved, reflected in enhanced technical 
writing skills such as the use of varied academic vocabulary, complex sentence structures, 
and improved inter-sentence coherence(Yogatama & Anggraheni, 2025).  

In contrast, students in the control group remained dependent on instructor 
corrections, resulting in writing that tended to be simplistic, repetitive, and underdeveloped. 
For example, below is a writing sample from one student in the experimental group on the 
health theme after using AI during drafting and revision: 

Student Writing Sample (Experimental Group, Health Theme): 

"Maintaining a healthy lifestyle is very important for university students. Regular 
exercise and balanced nutrition help to improve concentration and reduce stress. Using 

artificial intelligence applications, I can track my daily activities, such as calories intake 

and sleeping hours, which makes me more aware of my habits. In addition, AI also 

provides suggestions for better food choices and workout plans. Therefore, integrating 

technology in health management can support not only physical well-being but also 

academic performance." 

This writing demonstrates improvement in academic vocabulary usage (concentration, 
nutrition, academic performance), more complex sentence structures, and better coherence 
among ideas. The student successfully connected health concepts with technology utilization 
an element rarely observed in their initial writings prior to AI implementation. During the 
learning process in the control group, observations indicated that students did show some 
progress, but the improvement remained relatively limited. Their writings tended to be 
simplistic, featuring limited vocabulary, predominantly simple sentence structures, and 
recurring grammatical errors (Aprillia Wahyudi, 2025).  

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

341 
 

From a metacognitive perspective, students were unable to independently identify 
errors in their writing due to the absence of instant feedback. Meanwhile, regarding 
psychomotor aspects, technical writing skills developed only at a basic level using 
elementary vocabulary and simple paragraph structures, without significant variation. 
Writing motivation was also generally lower compared to the experimental group, as the 
learning process felt more passive and reliant on manual corrections from 
instructors(Rahayu Balai Diklat Keagamaan Ambon & Laksdya Leo Wattimena, n.d.-a). 

Student Writing Sample (Control Group, Health Theme): 

"Health is important for students. We must eat food and drink water every day. 

Exercise is good for body. Sleep is important because without sleep we cannot study 

well. Students must be healthy to study better and to have good life." 

This writing shows that although students understood the health theme, their 
presentation remained confined to simple, repetitive sentence patterns. Vocabulary usage 
lacked variety (food, water, sleep, study, good life), and ideas were not developed in depth. 
Inter-sentence coherence was also weak, making the writing resemble a list of points rather 
than a coherent academic paragraph(Rahayu Balai Diklat Keagamaan Ambon & Laksdya Leo 
Wattimena, n.d.-b). From this comparison, it is evident that the experimental group 
demonstrated significant progress in both metacognitive aspects (evaluating and revising 
writing) and psychomotor aspects (technical writing with varied vocabulary and complex 
structures). In contrast, the control group exhibited only moderate development at a basic 
level, without meaningful leaps in writing quality. 

CONCLUSION 

Based on the results of the research, it can be concluded that AI integration in writing 
learning provides impact significant positive in students’ writing ability.  Improvement high 
score in the group experiment show that AI is capable increase quality of writing through 
bait come back instant, adjustment materials, and support in the revision process. In 
addition, AI helps student improve grammar, expand vocabulary, as well as organize ideas 
more structured. AI also supports learning independent with give consistent and 
personalized guidance. The effectiveness of AI is very noticeable in improving motivation 
learning, trust self, as well as involvement active student in the learning process. With thus, 
AI is not only as tool help, but also as partners learning that can be speed up achievement 
competence writing as expected  

ACKNOWLEDGMENT 

The researcher would like to express sincere gratitude to the Ministry of Research and 
Technology/National Research and Innovation Agency Kemdiktisaintek for the generous 
support and funding provided through the PDP Research Grant, which made this research 
possible. Special appreciation is also extended to the Rector of Institut Ilmu Kesehatan (IIK) 
Pelamonia Kesdam XIV/Hasanuddin, for her continuous encouragement, guidance, and 
institutional support throughout the research process. 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

342 
 

The researcher is also indebted to colleagues, students, and all parties who contributed 
their time, effort, and valuable insights, which greatly enriched the quality of this research. 
Without their cooperation and assistance, this research entitled “The Influence of Artificial 
Intelligence Integration on Students’ English Writing Competence” would not have been 
successfully completed. 

REFERENCES 

Ahmed, M. R., Mekheimer, M. A., & Mahdi, E. (2025). Chatbots for Writing Skills: A Synthesis 

of Research. BSU-Journal of Pedagogy and Curriculum, 4(7), 254–275. 

https://doi.org/10.21608/bsujpc.2025.311761.1052 

Anderson, J. R., Corbett, A. T., Koedinger, K. R., & Pelletier, R. (1995). Cognitive tutors: Lessons 

learned. The Journal of the Learning Sciences. 4(2), 167–207. 

Aprillia Wahyudi, N. (2025). Analisis Penggunaan Media Belajar AI Untuk Meningkatkan 

Keterampilan Menulis Bahasa Inggris (Vol. 4). 

Carr, R., Palmer, S., & Hagel, P. (2015). Active learning: The importance of developing a 

comprehensive measure. Active Learning in Higher Education, 16(3), 173–186. 

https://doi.org/10.1177/1469787415589529 

Dana R. Ferris. (2010). SECOND LANGUAGE WRITING RESEARCH AND WRITTEN 

CORRECTIVE FEEDBACK IN SLA. Studies in Second Language Acquisition, 10(3), 181–

121. 

Fikri Amir Anggeraja, M., & Aeni, N. (2024). The Use of ChatGPT in Enhancing Students 
Writing Ability. Journal of Excellence in English Language Education, 3(3). 

Flower, L., & Hayes, J. R. (1981). A Cognitive Process Theory of Writing. College Composition 

and Communication, 32(4), 365. https://doi.org/10.2307/356600 

GHEDIR, H., & BOUCHAREB, A. (2025). The Role of Artificial Intelligence in Educational 

Practices for English as a Foreign Language (EFL) University Education: A Systematic 

Literature Review. ATRAS Journal, 6(2), 101–115. 

https://doi.org/10.70091/atras/vol06no02.7 

Godwin, R. (2015). The evolving roles of language teachers: Trained coders, local 

researchers, global citizens. Language Learning & Technology, 19(1), 10–22. 

http://llt.msu.edu/issues/february2015/emerging.pdfhttp://llt.msu.edu/issues/feb

ruary2015/emerging.pdf 

Ken Beatty. (2013). Teaching and researching computer-assisted language learning (2nd ed., 

Vol. 3). Routledge. 

Mondol, M. (n.d.). Statistical Analysis & Report on Pre-test & Post-test II. 

https://doi.org/10.13140/RG.2.2.11241.97128 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689


  
 

Copyright © The Author(s) 
Vol.6, No. 4, October 2025 

e-ISSN: 2723-4126 
p-ISSN: 2776-8880 

 

343 
 

Pasaribu, T. A. (2024). Quasi-experimental research design in language learning. Journal of 

Applied Linguistics and Education. 5(1), 33–47. 

Phan, T. N. Le. (2023). Students’ Perceptions of the AI Technology Application in English 

Writing Classes. Proceedings of the AsiaCALL International Conference, 4, 45–62. 

https://doi.org/10.54855/paic.2344 

Prokhorova, Y., Gujrati, R., & Uygun, H. (2024). The use of AI Chatbots in higher education: 

the problem of plagiarism. Review of Artificial Intelligence in Education, 5, e031. 

https://doi.org/10.37497/rev.artif.intell.educ.v5i00.31 

Rahayu Balai Diklat Keagamaan Ambon, S., & Laksdya Leo Wattimena, J. (n.d.-a). 

Pemanfaatan Artificial Intelligence (AI) dalam Penulisan Artikel Ilmiah. 

Rahayu Balai Diklat Keagamaan Ambon, S., & Laksdya Leo Wattimena, J. (n.d.-b). 

Pemanfaatan Artificial Intelligence (AI) dalam Penulisan Artikel Ilmiah. 

Rashed, A., & Almohesh, I. (2024). AI Application (ChatGPT) and Saudi Arabian Primary 
School Students’ Autonomy in Online Classes: Exploring Students and Teachers’ 

Perceptions. In International Review of Research in Open and Distributed Learning (Vol. 

25). 

Rogers, J. (2019). The Routledge Handbook of Research Methods in Applied Linguistics. 

https://www.researchgate.net/publication/334250281 

Sahem, F. (2024). The Role of Artificial Intelligence in Language Learning Language. 

Sung, Y. T. (2016). The effect of integrating technology into language learning. Educational 

Research Review. Https://Doi.Org/10.1016/j.Edurev.2016.05.001, 68–84. 

Yogatama, A., & Anggraheni, D. (2025). Meningkatkan Speaking dan Writing Skill Mahasiswa 

Menggunakan Artificial Intelligence (AI) pada Kelas Bahasa Inggris Psikologi. In 

Indonesian Research Journal on Education Web Jurnal Indonesian Research Journal on 

Education (Vol. 5). 

Yuan, L., & Liu, X. (2025). The effect of artificial intelligence tools on EFL learners’ 

engagement, enjoyment, and motivation. Computers in Human Behavior, 162, 108474. 

https://doi.org/10.1016/J.CHB.2024.108474 

  

 

http://u.lipi.go.id/1593190689
http://u.lipi.go.id/1593190689

