JEELS
(Journal of English Education and Linguistics Studie.
P-ISSN: 2407-2575 E-ISSN: 2503-2194 https://jurnalfaktarbiyah.
id/index.
php/jeels STUDENT ENGAGEMENT WITH ARTIFICIAL INTELLIGENCE
(AI) CHATBOT IN ENGLISH LANGUAGE LEARNING: THEORY
OF PLANNED BEHAVIOR PERSPECTIVE
Fauziah Fauziah1.
Novita Diana2.
Silvia Putri3.
Teuku Fadhli4 1,2,3English Education Department.
Universitas Jabal Ghafur.
Aceh.
Indonesia.
4Guidance & Counseling Department.
Universitas Jabal Ghafur.
Aceh.
Indonesia *fziah05@yahoo.
novitadiana111@gmail.
silviaputri091217@gmail.
teuku_fadhli@unigha.
(*) Corresponding Author Abstract: This qualitative study investigated student engagement in English language learning facilitated by Artificial Intelligence (AI) chatbots, utilizing the Theory of Planned Behavior (TPB) Employing a narrative inquiry, this study explores the lived experiences and perceptions of 15 participants enrolled in a private higher education institution in Aceh.
Indonesia.
Data collection involved semi-structured interviews and focus group discussions to explore studentsAo attitudes, social influences, and perceived behavioral control.
Thematic analysis was applied to analyze the data, revealing studentsAo attitudes Citation in APA style:
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
DOI: 10.
30762/jeels.
Submission: December 2024.
Revision: April 2025.
Publication: June 2025 Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
towards AI chatbots influenced by perceptions of utility and social norms.
Participants had to deal with both their personal views and things happening around them, like support from others or access to technology, as they used AI chatbots to help them learn English.
This study emphasizes the importance of considering sociocultural contexts and user experience in the design and implementation of AI-driven educational tools.
The findings support previous research highlighting the importance of user attitudes, social influences, and perceived control in shaping students' engagements with technology in educational contexts.
The insights gleaned from this research contribute to the broader discourse on technology-mediated language learning and inform strategies for enhancing student engagement in English language education.
Keywords: AI chatbot, narrative inquiry, student engagement.
English learning INTRODUCTION The integration of Artificial Intelligence (AI) into education has experienced steady growth in recent years, transforming the ways in which students engage with learning materials across all educational levelsAifrom early childhood to higher education (Crawford.
Allen.
Pani, & Cowling, 2024.
Ng et al.
, 2023.
Yang, 2.
AI technologies such as machine learning, natural language processing, and intelligent tutoring systems have enabled more personalized and adaptive learning experiences.
In particular, the domain of language learning has witnessed notable advancements, with AI-powered applications offering individualized support, immediate feedback, and immersive practice environments tailored to learners' needs.
Kong et al.
underscore the increasing importance of AI literacy within educational contexts, emphasizing its relevance for future workforce preparedness.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
informed decision-making, ethical awareness, critical thinking, and They advocate for a comprehensive educational approach that integrates both technical competencies and critical, ethical, and social dimensions of AI, highlighting the need for collaboration among educators, policymakers, and other stakeholders in developing effective AI literacy curricula.
Building on this development, one of the most prominent applications of AI in language education is the use of chatbots.
These AI-driven tools are designed to simulate human conversation, thereby offering learners an interactive and dynamic platform for practicing language skills in real time (Liu & Ma, 2.
With ongoing advancements in AI algorithms, chatbots are increasingly capable of understanding and responding to user input with a high degree of accuracy, positioning them as effective virtual tutors.
Existing studies suggest that AI chatbots can facilitate a variety of language learning tasks, including vocabulary enrichment, grammar correction, and conversational practice, all of which contribute to a more engaging and individualized learning experience (Polakova & Klimova, 2024.
Liu et , 2.
Moreover, research highlights several pedagogical benefits associated with the use of AI chatbots in English language learning.
First, they offer a non-judgmental and patient practice partner, which can be particularly advantageous for learners who experience anxiety or hesitation when speaking in front of others.
Second, the 24/7 accessibility of AI chatbots allows learners to engage with the language at their own pace and convenience, thereby supporting consistent practiceAian essential component of effective language acquisition.
Third, these tools are capable of adapting to individual learner needs by providing personalized feedback and tailoring learning pathways based on users' progress and performance (Polakova & Klimova, 2024.
Liu et al.
, 2.
These features demonstrate the potential of AI chatbots Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
to enhance learner engagement and support more effective language learning outcomes.
Despite the growing adoption of AI chatbots in educational settings, particularly language learning, there is a notable gap in the qualitative understanding of how these tools affect student While numerous studies have quantitatively assessed the effectiveness of AI chatbots in improving language skills, there is limited qualitative research exploring the nuanced experiences and perceptions of students who use these tools.
Existing research tends to focus on measurable outcomes such as test scores, vocabulary acquisition, and grammatical improvements.
However, these studies have often overlooked the subjective and contextual factors that influence student engagement, such as individual attitudes towards AI chatbots, the role of social influences, and students' perceived control over their learning processes.
Engagement is a multifaceted construct that encompasses emotional, cognitive, and behavioral dimensions, and it is crucial to understand how AI chatbots impact these different aspects from students' perspectives (Lin & Yu, 2025.
Zou.
Lyu.
Han.
Li, & Zhang, 2.
Theory of Planned Behavior (TPB) provides a valuable framework for investigating these factors, suggesting that behavior is influenced by attitudes, subjective norms, and perceived behavioral controls (Ajzen, 1.
However, there is a small number of researchers that applies PBT to the context of AI-driven language learning.
This gap limits our understanding of how students' beliefs and social environments shape their interactions with AI chatbots and how these interactions affect their engagement and learning outcomes.
This study addresses this gap by qualitatively exploring students' lived experiences and perceptions of using AI chatbots in English language By focusing on the subjective experiences of learners, this study aims to uncover the motivational and contextual factors that drive engagement.
Understanding these factors can lead to the Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
development of more effective AI chatbots that are better aligned with students' needs and preferences, ultimately enhancing their learning experiences and outcomes in English language education.
Given this backdrop, the purpose of this study is to explore students' lived experiences and perceptions concerning the integration of AI chatbots into English language learning.
To address this aim, this study poses the following research question:
How do students articulate their attitudes towards using AI chatbots for English language learning? What social influences .
ubjective norm.
impact students' engagement with AI chatbots? How do students construe their control over using AI chatbots in language learning? This study aimed to understand students' personal evaluations and emotional responses to AI chatbots, examine the social influences that affect their engagement, and analyze their perceived ability to use these tools effectively.
Furthermore, this study seeks to offer a comprehensive account of how these factors .
ttitudes, subjective norms, and perceived behavioral contro.
manifest in students' engagement with AI chatbots, encompassing emotional, cognitive, and behavioral dimensions.
The study also intends to draw practical implications for educators, developers of AI chatbots, and policymakers, providing recommendations for enhancing student engagement and learning outcomes through the effective integration of AI chatbots in language education.
By addressing these objectives, this study aims to fill the existing gap in qualitative research on AI chatbots in English language learning, offering valuable insights into the factors that influence student engagement, and laying the groundwork for improving the design and implementation of AIdriven educational tools.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
METHOD
Research Design A narrative inquiry approach was adopted to explore studentsAo lived experiences with AI chatbots.
Narrative inquiry is particularly suited for capturing the complexities of engagement behaviors and individual contexts, providing rich, qualitative insights (Clandinin & Connelly, 2.
The study employs TPB as a guiding framework to interpret these narratives and connect them to broader behavioral patterns.
Participants Participants were recruited from English language courses at one of a higher education in Aceh.
Indonesia, specifically those who had prior experience using AI chatbots for language learning.
purposive sampling technique was used to ensure that selected participants had relevant experiences to contribute to the studyAos This approach ensured that the sample was relevant and could provide rich, detailed data about their interactions with the technology.
Ultimately, this study involved 15 students who were actively enrolled in English language courses in a private University in Aceh.
Indonesia.
The study was conducted over a period of 10 months, from August 2023 to May 2024, encompassing two academic semesters.
One of the researchers also participated as a lecturer in participantsAo daily This involvement allowed for a deeper understanding of the classroom dynamics and provided firsthand experience with the integration of AI chatbots in the learning process.
All participants were provided with detailed information about the study and gave informed consent before participating.
Confidentiality was maintained throughout the study, and data were pseudonym to protect participants' identities .
ee Table .
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
Table 1.
ParticipantsAo demographic data Pseudonym Nisa Salsa Lita Sera Nana Zara Liz Ais Tria Lian Arai Randa Ama Gender Female Female Female Female Female Female Female Female Female Female Female Female Female Male Male Age Semester 2nd years 3th years 2nd years 3th years 2nd years 2nd years 3th years 2nd years 3th years 3th years 2nd years 2nd years 2nd years 3th years 3th years Data Collection Data were collected through semi-structured interviews and focus group discussions.
Semi-structured interviews were conducted to investigate individual students' experiences, attitudes, subjective norms, and perceived behavioral control related to the use of AI These interviews allowed for flexibility and depth, enabling participants to express their thoughts and feelings comprehensively (Seidman, 2.
Group discussions were organized to gather insights into shared experiences and the social dynamics influencing students' engagement with AI chatbots.
The interactive nature of focus groups facilitated the exploration of collective attitudes and norms (Krueger & Casey, 2.
A detailed interview guide comprising open-ended questions was developed based on the constructs of the Theory of Planned Behavior (TPB) and the dimensions of student engagement (Ajzen.
Miao.
Chang, & Ma, 2022.
Zou et al.
, 2.
The questions aimed to explore studentsAo personal experiences, attitudes, social influences, and perceived control in relation to AI chatbot use in language learning, as well as their emotional, cognitive, and behavioral Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
To ensure the reliability and clarity of the guide, the interview protocol underwent an expert validation process involving two lecturers with expertise in educational research and qualitative Their feedback was used to refine the wording, structure, and alignment of the questions with the theoretical constructs.
Additionally, a pilot interview was conducted with one participant who met the sampling criteria but was not included in the main study.
This pilot helped to identify any ambiguous or leading questions and allowed for minor adjustments to improve the flow and effectiveness of the interview.
These steps ensured that the guide was both theoretically grounded and practically applicable, enhancing the trustworthiness of the data collection process.
The interviews were conducted in private settings to encourage openness and honest reflection, while the focus group sessions were held during classroom reflection activities in a Technology in Language Teaching course.
Both formats followed a protocol designed to explore attitudes, subjective norms, and perceived behavioral control in line with the Theory of Planned Behavior (TPB) framework (Ajzen, 1991.
Miao et al.
, 2.
All interviews and discussions were conducted in Indonesian, the participantsAo formal academic language, to facilitate clearer expression of their thoughts.
With informed consent from each participant, all sessions were digitally recorded.
These recordings were transcribed verbatim to ensure data accuracy and richness for analysis.
Not all participants engaged in both data collection methods.
Individual interviews were conducted with a subset of participants to delve deeper into personal experiences, while focus groups provided insights into collective views and social dynamics.
Before participation, all students were briefed about the purpose of the study, the voluntary nature of their involvement, and their right to withdraw at any time without consequence.
Participants were assured that there would be no academic penalty or reward associated Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
with participation.
To protect their identities, pseudonyms were assigned during transcription and reporting, and all data were securely The study posed minimal risk, as the topics discussed were related to students' learning experiences.
however, participants were informed that they could skip any questions they were uncomfortable The potential benefits included contributing to the improvement of language teaching practices through AI-enhanced tools.
Data Analysis Thematic analysis served as the primary methodological approach to analyze qualitative data collected through semi-structured interviews and focus groups in this study on student engagement with AI chatbots in English language learning contexts.
According to Braun and Clarke .
, thematic analysis steps .
ee Figure .
involves systematically identifying patterns and themes within data to gain insights into participants' experiences and perceptions.
Transcripts from interviews and focus groups undergone rigorous coding to extract themes related to students' attitudes, subjective norms, and perceived behavioral control concerning AI chatbots.
This method facilitated the exploration of how these factors influence engagement with technology-enhanced learning tools.
The example of data analysis can be seen in Table 2.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
Figure 1.
Braun and Clarke .
Ao thematic analysis Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
Table 2.
The process of thematic analysis Braun and Clarke .
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
FINDINGS
This study investigated the extent and nature of student engagement with AI chatbots specifically designed for English language learning.
The findings presented include: .
studentsAo attitudes towards AI chatbots, .
the significant influence of social norms on students engagement with AI chatbots, and .
Perceived control over using AI chatbots.
The findings reveal nuanced insights into how students interacted with and perceived the AI chatbot within the context of language education.
StudentsAo Attitudes towards Using AI Chatbots Findings from this study reveal a spectrum of attitudes towards AI chatbots among students engaged in language learning.
These attitudes range from enthusiasm and appreciation for the chatbots' ability to provide immediate feedback and personalized learning experiences, to concerns regarding their reliability, particularly in handling complex language tasks.
The diversity in these attitudes underscores the critical importance of considering individual preferences and needs when designing and implementing AI-driven educational tools.
Such considerations are vital for effectively enhancing student engagement and optimizing the educational benefits of AI technology in language learning contexts.
The following excerpts provide a glimpse into how students perceive and interact with the AI chatbot in different contexts, reflecting their attitudes, behaviors, and experiences during language learning activities.
AuI find the chatbot really helpful because it gives me instant feedback on my pronunciation.
I can practice speaking without feeling embarrassed in front of the class.
(Nisa, 07 Dec, intervie.
Using the chatbot together in the class has been fun.
We challenge each other with different questions, and it feels like a game Ay (Salsa, 05 Dec 2023.
FGD) During the FGD, students shared their experiences using the AI chatbot independently in the classroom.
They appeared engaged, with Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
many actively typing responses and repeating phrases aloud to practice pronunciation.
Some students paused to read the chatbot's explanations before continuing with the next exercise.
Despite her high confidence in using AI chatbots, one of the participants.
Nisa, identified usability factors like advertisements during extended use that may impact sustained engagement: AuSometimes ads pop up while practicing, so itAos a bit distracting and I have to close them firstAy.
Another participant.
Mira, commented that some features were locked behind a paywall: AuThe better features, like pronunciation correction, are only available for premium usersAy.
While acknowledging the chatbotAos capability for translation.
Ais noted occasional inconsistencies: AuItAos good at translating individual words, but sometimes long sentences become weird or disconnectedAy.
These FGD notes suggest that although students appreciated the chatbot's assistance, they also evaluated its limitations.
Several students also expressed concerns about the reliability of the information provided.
For instance.
Farah stated.
AuSometimes its answers differ from what the lecturer taught, so I need to double-check in the book or ask friendsAy.
The Influence of Social Norms on StudentsAo Engagement Participants' experiences emphasize the significant influence of social norms on their engagement with AI chatbots.
Peer influence emerges as a strong motivating factor, with participants feeling compelled to use the chatbot due to the popularity and endorsement of their peers or teachers.
These findings align with existing literature on the role of social factors in shaping technology adoption and usage behaviors among students (Chou et al.
, 2023.
Miao et al.
, 2022.
Soland et al.
, 2.
The data from semi-structured interviews and group discussions indicate that social norms significantly influence students' engagement with AI chatbots.
The findings reveal that peer influence is a critical motivating factor, with participants frequently citing the popularity and endorsement of their peers as primary reasons for their adoption and use of the chatbot.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
The interviews provide detailed insights into how social factors impact individual decisions to use AI chatbots.
For instance.
Nana highlighted peer influence as a major motivator, stating:
AuHonestly, it was because my friends were all talking about it and using it.
They kept saying how helpful it was for their studies, so I decided to give it a try.
Ay (Nana, 05 Dec 2.
Similarly.
Zara emphasized the role of social influence:
AuOne of my friends said that she used perplexity to help her in doing assignment.
Seeing my classmates using it and sharing their positive experiences, it made me curious and more open to trying it myselfAy.
(Zara, 14 Dec 2.
These responses underscore the importance of social validation and the desire to conform to group behaviors, aligning with existing literature on the social determinants of technology adoption (Zhao et , 2.
Group discussions further illuminate the communal aspects of AI chatbot adoption.
In one discussion, participants explicitly linked their usage of the chatbot to peer behaviors:
AuFor me, it was seeing how much it helped my friends with their They were always talking about it, so I felt like I should use it too.
Ay (Liz, 18 Dec 2.
AuItAos like, when everyone around you is using it and benefiting from it, you donAot want to miss out.
Ay (Nana, 18 Dec 2.
These comments suggest a strong social component where peer usage not only sparks interest but also reinforces the perceived value of the chatbot through shared experiences.
This finding supports the notion that social influence and peer pressure can significantly drive technology adoption, as suggested by social influence theory (Chou et , 2023.
Miao et al.
, 2022.
Soland et al.
, 2.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
An insightful moment occurred during the FGD session when the first author revisited the use of the EnglishScore app .
n AI-supported English proficiency testing tool previously introduced to the Although not part of the formal research instruments, this activity served to review studentsAo familiarity and engagement with AI-enhanced language learning technologies.
As participants reengaged with the app, they encountered challenges navigating certain features, which prompted spontaneous collaborative behaviors.
Students grouped together to troubleshoot issues, demonstrate functionalities, and support one anotherAos understanding of the app.
This peer-assisted learning dynamic highlighted aspects of behavioral and social engagement, reinforcing the studyAos broader The participantsAo willingness to help one another, their shared problem-solving, and their persistence in working through technical obstacles aligned with the constructs of subjective norms and perceived behavioral control within the Theory of Planned Behavior (TPB).
Although the EnglishScore app was not the central focus of the study, this activity provided a valuable contextual lens through which to understand students' real-time interaction with AI-based tools in language learning settings.
During the FGD session, students were discussing their experiences with the EnglishScore App.
Peer recommendations and demonstrations were common, with students showing each other how to use certain features.
Dec 2.
In classroom settings, peer support played a crucial role in the initial adoption phase:
Students who used the EnglishScore app were frequently observed helping their peers set up and navigate the EnglishScore interface.
This peer support seemed to play a crucial role in the initial adoption among students.
Dec 2.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
Perceived Control over Using AI Chatbots The data reveal varying levels of perceived control over using AI chatbots among participants.
Technological proficiency and time management are identified as key factors influencing participants' perceived ease or difficulty in using the technology.
These findings align with previous research highlighting the importance of userfriendliness and accessibility in facilitating technology acceptance and adoption (Lau.
Qian, & Chiu, 2025.
Liu et al.
, 2.
Participants' comments during the interviews illustrate the impact of technological proficiency on their perceived control.
For example.
Sera, who identified as technologically proficient, expressed confidence in using the chatbot:
:IAove always been good with tech, so using the chatbot was pretty straightforward for me.
I didn't face any major issues, and it was easy to integrate it into my routine.
Ay (Sera, 29 April, 2.
In contrast.
Ama, who reported lower technological proficiency, described encountering difficulties:
AuI'm not very tech-savvy, so I struggled a bit with getting the chatbot to work properly.
It took me a while to figure out how to use it effectively.
Ay (Ama, 12 Aug, 2.
These differing experiences underscore the importance of technological proficiency in shaping users' perceived control and ease of use.
This supports the technology acceptance model, which posits that perceived ease of use significantly impacts users' acceptance of new technologies (Miao et al.
, 2022.
Xue.
Rashid, & Ouyang, 2.
Group discussions highlighted the role of time management in participants' experiences with the AI chatbot.
Participants who managed their time well found the chatbot to be a useful tool, while those with poor time management skills faced challenges.
For example:
AuI usually plan my study sessions in advance, so using the chatbot fit nicely into my schedule.
It actually helped me save time on research and get quick answers.
Ay (Lian, 28 Nov 2.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
On the other hand:
AuI have a hard time managing my time, and trying to learn how to use the chatbot just felt like another thing on my plate.
It is hard to focus on phone coz I sneak and scroll my social media in the same It was more stressful than helpful at times.
Ay (Ly, 28 Nov 2.
These discussions highlight that effective time management can enhance the perceived utility of the chatbot, while poor time management can exacerbate difficulties, supporting findings from previous research on the role of self-regulation in technology use (Chou et al.
, 2.
Other participants provided further evidence of the impact of technological proficiency and time management on perceived control.
Technologically proficient students were often seen navigating the chatbot effortlessly and helping peers who encountered technical AuTechnologically proficient students were frequently assisted their peers with technical issues related to the chatbot.
They appeared more confident and efficient in using the technology.
Ay (Randa, 09 Jan 2.
Conversely, the FGD note highlights significant user frustration with the chatbot app due to excessive and intrusive advertisements that disrupt focus and overshadow the app's primary function.
The ads are more engaging than the chatbot itself, leading the user to lose interest and abandon the app.
Poor navigation within the app exacerbates the issue, as users struggle to avoid being redirected by the This point to a need for a more user-centric design, reducing advertisement intrusiveness, enhancing chatbot functionality, and simplifying the appAos interface to improve user retention and AuI found the chatbot experience frustrating, leading me to abandon its use after just a few initial attempts.
The application is Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
overwhelmed with advertisements that constantly interrupt my focus and detract from the main purpose of using the chatbot.
Ironically, the advertisements are often more captivating than the chatbot itself, which causes me to lose track of my original intent.
Additionally, the appAos design makes it easy to get lost navigating through these ads, further diminishing its usability and appeal.
Ay (Arai, 12 Jan 2.
These data reinforce the notion that effective time management are crucial for fostering a sense of control and ease in using AI chatbots.
This aligns with the unified theory of acceptance and use of technology, which emphasizes the importance of facilitating conditions and selfefficacy in technology adoption (Alemayehu & Chen, 2023.
Cai.
Lin, & Yu, 2024.
Soland et al.
, 2.
Students' Perceptions and Actual Engagement The findings suggest a complex interplay between students' perceptions and their actual engagement with AI chatbots.
Positive attitudes and perceptions of utility often lead to consistent engagement, while concerns about reliability and technical issues can hinder usage.
Additionally, social norms and peer dynamics play a significant role in shaping students' engagement patterns, highlighting the need for a holistic understanding of the factors driving technology usage behaviors.
The interviews reveal how students' perceptions of utility and reliability influence their engagement with AI chatbots.
For instance.
Arai expressed a positive attitude towards the chatbot, which translated into regular usage:
AuI find the chatbot really useful for quick answers and study tips.
It's like having a tutor available every time everywhere.
Because of this.
I use it almost every day.
Ay (Arai, 19 Dec 2.
In contrast.
Ais highlighted concerns about reliability, which affected their engagement:
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
AuSometimes the chatbot gives me answers that donAot make sense or are too vague.
This makes me hesitant to rely on it, so I only use it occasionally.
Ay (Ais, 02 Jan 2.
These responses indicate that while positive perceptions of utility can enhance engagement, concerns about reliability can deter consistent usage.
This supports the technology acceptance model, which emphasizes the impact of perceived usefulness and perceived ease of use on technology adoption (Chiu, 2024.
Lau et al.
, 2.
Group discussions shed light on how social norms and peer dynamics influence students' engagement with AI chatbots.
For example.
Tria described how peer recommendations enhanced their AuEveryone in my study group uses the chatbot and we often share tips on how to get the best answers from it.
This has made me more inclined to use it regularly.
Ay (Tria, 19 Dec 2.
Conversely.
Lian noted that negative peer feedback affected their AuA few of my friends had bad experiences with the chatbot, like getting incorrect answers.
Listening about their issues made me wary of using it too much.
(Lian, 02 Jan 2.
These discussions highlight the significant role of social influence in shaping technology usage behaviors.
Positive peer dynamics can promote engagement, while negative experiences shared among peers can discourage use.
This aligns with the social influence theory, which posits that individualsAo behaviors are influenced by the expectations and behaviors of their social group (Fabiyi 2024.
Chou et al.
, 2.
FGD data provide further insights into the factors affecting engagement with AI chatbots.
For instance, students with positive perceptions of the chatbot's utility were observed using it frequently and integrating it seamlessly into their study routines:
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
During the discussion, several students shared their perspectives on the chatbot's utility as a study aid.
Students who perceived the chatbot as beneficial for their learning process were observed using it consistently during study sessions.
They frequently asked questions related to their coursework, explored various features, and leveraged the tool to supplement their understanding of challenging topics.
This group expressed satisfaction with the chatbotAos ability to provide quick responses and additional resources, which they found helpful in improving their academic .
Jan 2.
On the other hand, students who encountered technical issues or questioned the chatbot's reliability were observed engaging less Several students expressed frustration when the chatbot provided unclear or incorrect answers, leading them to reduce their usage over time.
Jan 2.
Additionally, the influence of peer interactions was evident:
In group settings, students who received positive feedback from peers about the chatbot were more likely to use it actively.
Conversely, negative feedback from peers resulted in visibly reduced engagement.
Jan 2.
These FGD notes underscore the importance of perceived utility, reliability, and social dynamics in determining students' engagement with AI chatbots.
DISCUSSION
The first finding reflects a degree of critical thinking and selfmonitoring in the learning process.
The ability of students to critically evaluate disciplinary information obtained from AI chatbots is essential (Chiu, 2.
Developing critical reasoning and thinking skills is a prerequisite for students to engage effectively with chatbots in their learning process.
These skills enable them to discern the quality and reliability of AI-generated content, ensuring a more informed use of Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
such tools.
As Tlili et al.
highlight, students must learn to critically assess and curate chatbot responses, and educators have a key role in nurturing these competencies.
Participants' responses in interviews and focus group discussions highlight a dual perception of the AI chatbot as both a beneficial learning tool and a source of concern.
All of 15 participants expressed enthusiasm for the chatbot's immediate feedback and personalized learning experiences, noting its ability to enhance language practice without fear of judgment.
For instance.
Nisa described the chatbot as "helpful" for improving pronunciation skills and providing a safe space for practice.
On the contrary, concerns were voiced regarding the chatbot's reliability, particularly in handling complex language tasks.
Salsa mentioned during a group discussion that while using the chatbot together was enjoyable, occasional misunderstandings occurred, which could be frustrating.
This sentiment was echoed in FGD notes, where instances of students pausing to review and rephrase responses suggested occasional challenges in understanding and communicating effectively with the These insights underline the importance of balancing the technological benefits of AI chatbots with considerations of their practical limitations.
Understanding these dynamics is crucial for designing AI-driven educational tools that effectively support and enhance student engagement in language learning contexts.
Participants' varied attitudes towards the AI chatbot align closely with the constructs of Planned Behavior Theory (TPB), which posits that attitudes, subjective norms, and perceived behavioral control influence behavioral intentions and subsequent behaviors.
The positive attitudes expressed towards the chatbotAos immediate feedback and personalized learning experiences reflect favorable perceptions of its utility and effectiveness in enhancing language practice.
These perceptions align with TPB's emphasis on the positive evaluation of behavior as a determinant of behavioral intention (Ajzen, 1.
This is consistent with findings by Polakova and Klimova .
, who observed that learners appreciated the immediate, adaptive responses Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
from AI chatbots, which promoted learner autonomy and boosted Similarly.
Crawford et al.
found that chatbot feedback contributed to students' perceived academic support and However, concerns about the chatbot's reliability in handling complex or nuanced language tasks indicate limitations in perceived efficacy, which can reduce usersAo sense of control.
According to Ajzen .
, perceived behavioral control encompasses the perceived ease or difficulty of performing a behavior, influenced by factors such as technical reliability and user proficiency.
These findings resonate with FabiyiAos .
study, where students expressed frustration when chatbot responses lacked contextual understanding, leading to reduced confidence in the tool.
In this study, such concerns were also evident.
Nisa.
Mira, and Ais expressed frustration during group discussions when the chatbotAos answers did not align with their expectations, especially regarding locked features or inconsistent translations.
This reflects the gap between anticipated and actual performance, a phenomenon observed in Meyer von Wolff et al.
, who found that student engagement declines when chatbot functionality fails to meet pedagogical needs.
The FGD notes and classroom interactions further support these findings, documenting moments where students paused to rephrase their responses or seek clarification, signaling attempts to overcome perceived interaction barriers.
These observations mirror Chen et al.
findings, which emphasized the need for improved natural language processing (NLP) to support more authentic, fluid communication in language learning contexts.
The second findings focus on the role of peer facilitation and collaborative learning environments in promoting the use of AI This aligns with the diffusion of innovations theory (Lau.
Qian, & Chiu, 2025.
Miao, 2.
, which posits that interpersonal networks are vital in the spread of new technologies.
Understanding the diffusion process can help in designing better strategies for promoting and implementing innovations effectively.
This experience Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
underscored the practical benefits of integrating digital tools into the learning environment, as well as the necessity of providing initial guidance and support (Li et al.
, 2024.
Soland et al.
, 2.
The students' engagement and adaptability underscored the value of creating opportunities for them to collaborate and develop digital literacy skills.
This FGD reinforces the importance of incorporating peer-assisted learning strategies in educational settings to enhance students' critical thinking and problem-solving abilities.
The convergence of findings shows the significant influence of social norms and peer interactions on the adoption and use of AI chatbots among students.
Peer influence emerges as a powerful factor, with students feeling compelled to use the chatbot due to its popularity and the positive endorsements from their peers.
This studyAos results are consistent with existing literature that underscores the role of social factors in technology adoption.
The desire to conform to group behaviors, seeks social validation, and benefit from communal learning experiences are key drivers behind the engagement with new technologies like AI chatbots (Goh, & Sigala, 2.
These findings suggest that educational institutions aiming to promote the use of AI chatbots should consider strategies that leverage peer influence and foster collaborative learning environments.
A multifaceted relationship between students' perceptions and their engagement with AI chatbots shows positive attitudes and perceived utility often result in regular usage, while concerns about reliability and technical issues can impede consistent engagement.
Social norms and peer dynamics further complicate this relationship, underscoring the need for a comprehensive understanding of the factors influencing technology usage behaviors.
These results are consistent with the technology acceptance model and social influence theory, which emphasize the roles of perceived usefulness, perceived ease of use, and social factors in technology adoption.
Educational institutions aiming to enhance AI chatbot engagement should address reliability concerns, enhance user training, and leverage positive peer dynamics to foster a supportive environment for technology use.
Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
The findings from interviews and group discussions collectively indicate that perceived control over using AI chatbots varies significantly among participants.
Technological proficiency and time management emerge as pivotal factors influencing this perceived Participants who are more technologically savvy exhibit greater confidence and ease in using the chatbot, whereas those with lower proficiency and poor time management skills encounter significant challenges.
These findings corroborate existing literature on technology acceptance, highlighting the critical role of userfriendliness and accessibility in technology adoption.
The data suggest that educational institutions aiming to promote AI chatbot usage should consider implementing training programs to enhance technological proficiency and provide resources to support effective time management among students.
The findings of this study contribute to the existing literature on technology-enhanced language learning and the adoption of AI-driven educational tools.
They corroborate previous research highlighting the importance of user attitudes, social influences, and perceived control in shaping students' engagement with technology in educational Moreover, the study provides nuanced insights into the unique affordances and challenges of AI chatbots in language learning, adding to our understanding of their potential impact on student learning outcomes.
CONCLUSION
This qualitative study examined student engagement with an AI chatbot designed for English language learning through the lens of Planned Behavior Theory (PBT).
By employing a phenomenological approach, the research explored the lived experiences and perceptions of 15 participants from a private higher education institution in Aceh.
Indonesia.
Through two different instruments .
emi-structured interviews and focus group discussion.
, attitudes, social influences, perceived behavioral control, and the alignment of perceptions with engagement were examined.
Thematic analysis of the data revealed a Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
diverse spectrum of attitudes towards AI chatbots, shaped by perceptions of utility, social norms, and technological proficiency.
The findings from interviews and focus groups highlighted converging themes and provided a comprehensive understanding of how students engage with AI chatbots in English language learning contexts.
This study emphasizes the significance of considering sociocultural contexts and user experience in the design and implementation of AI chatbots for educational purposes.
underscores the need for tailored strategies that address both the technical capabilities of AI chatbots and the socio-behavioral dynamics influencing student engagement.
By integrating insights from Planned Behavior Theory, educators and developers can enhance the effectiveness of AI-driven tools in promoting student engagement and learning outcomes in language education.
Nonetheless, this study is not without limitations.
The relatively small sample size and single-institution focus may restrict the generalizability of the findings to broader populations.
Additionally, the reliance on self-reported data through interviews and focus groups may be influenced by participantsAo biases or memory recall.
The study also did not explore long-term engagement or the effectiveness of chatbots on actual language performance, which could be areas for future investigation.
Future researchers are encouraged to conduct longitudinal studies with more diverse and larger populations to validate and expand upon these findings.
Mixed-method approaches that incorporate quantitative data .
uch as engagement analytics or language assessment score.
could offer a more holistic view of the chatbotAos impact.
Further exploration of specific design features .
personalization, gamificatio.
and their influence on engagement could also inform more effective chatbot development for educational DECLARATION OF AI AND AI-ASSISTED TECHNOLOGIES During the preparation of this work the authors made use of generative AI tools for Theoretical Development .
heorem proving, conceptual analysis.
Editing .
rammar, readabilit.
, and Citation Fauziah.
Diana.
Putri.
, & Fadhli.
Students engagement with Artificial Intelligence (AI) Chatbot in English language learning: Theory of planned behaviour perspective.
JEELS, 12.
Formatting .
tructuring, organizing, optimizing language and The authors have reviewed and edited the content as needed and take full responsibility for the content of the publication after the use of this tool/service.
REFERENCES