Perceptions of pre-service teachers on the use of ChatGPT during their training and the future implications for their future role in high schools: A research in progress

doi:10.21203/rs.3.rs-4542918/v1

Download PDF

Case Report

Perceptions of pre-service teachers on the use of ChatGPT during their training and the future implications for their future role in high schools: A research in progress

https://doi.org/10.21203/rs.3.rs-4542918/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The reception of ChatGPT in the educational field has varied between enthusiasm and scepticism, generating a series of controversies and challenges in response to the emergence of artificial intelligence (AI) technologies within the educational context. The purpose of this study was to interpret and describe the perceptions of a group of pre-service teachers on the use of ChatGPT as an assistant in a learning experience implemented during their training, and on the implications of this AI tool in their future work context as teachers. A sequential mixed methodology was employed, initially collecting quantitative data, followed by a deeper analysis based on qualitative data from which explanatory categories of the phenomenon emerged. Our study on the integration of ChatGPT into teacher training programs reveals that students valued the use of ChatGPT for its conceptual support and quick feedback, enhancing their learning experience. However, concerns about the accuracy of information and plagiarism underline the need for a critical engagement with AI tools. Regarding their future teaching role, they highlighted the importance of mediation, training, and guiding students in the critical and responsible use of AI. They also identified the need to modify assessment methods to encourage critical thinking, problem-solving, and creativity. The study emphasized the importance of integrating AI technologies into teacher training in a reflective and critical manner, with guidance from educators and IT specialists. This These findings suggest a balanced approach to incorporating AI in education, emphasizing ethical use and the transformative potential of AI to support pedagogical objectives.

pre-service teacher

ChatGPT

critical thinking

academic integrity

artificial intelligence (AI)

student perceptions

ethical issues

In the realm of higher education, the integration of innovative AI technologies has emerged as a new challenge in the context of training future educators. Facing the changes and advancements presented by AI in 21st-century educational settings, it is necessary for students to develop skills to interpret, analyse, and ethically engage with the information provided by these models, thereby strengthening their digital competencies [1, 2]. Current AI tools like ChatGPT, being more sophisticated models, exhibit advanced natural language processing capabilities to handle information from multiple sources or datasets. This enables them to respond to issues or generate complex content under specific requirements and in a personalized manner [3]. For this reason, ChatGPT can be utilized in various ways in teaching and learning processes, directly impacting the pedagogical practices of future teachers.

The use of ChatGPT in higher education offers diverse opportunities, from providing writing assistance and assessment feedback to personalized guidance, thereby enhancing student engagement and motivation [3–5]. The integration of ChatGPT presents an opportunity to prepare pre-service teachers who are interacting in real educational environments and participating in discussions about the uses and challenges of AI in secondary schools.

The integration of ChatGPT into educational environments also raises ethical considerations, requiring a critical analysis of its uses and challenges, particularly in the context of training future educators [3, 6]. Concerns regarding academic integrity, plagiarism, and the responsible use of AI technologies underscore the importance of establishing clear ethical procedures and norms tailored to the context and the students' training needs [7]. It is also deemed essential to prevent plagiarism and promote responsible AI use by guiding its application in student assistance tasks and more authentic assessments that go beyond mere information or knowledge reproduction [8].

For educators to integrate AI technologies into their teaching, they need to possess the skills and experiences to do so, serving as a foundation for developing technological competencies for the use of new AI technologies. Amid these challenges, there is a limited and scant body of research addressing future teachers' perceptions in this area, through applied experiences during their training, which promote concrete student interactions with AI, as well as critical reflection on its responsible and ethical use in their future teaching performance [9].

With this study, we aim to contribute to the effective and ethical integration of ChatGPT in higher education, with a particular emphasis on preparing pre-service teachers to leverage the opportunities presented by AI technologies and guide students in developing critical thinking skills, through the responsible use of ChatGPT in their learning processes [10].

This research hopes to add to the current literature on the subject, from a South American perspective, providing valuable insights from the voices of pre-service teachers as active participants in exploring the use of ChatGPT during their training in a course at a Chilean university. The goal is to interpret and describe the perceptions of a group of pedagogy students on the use of ChatGPT as an assistant in a learning experience and on the implications of this tool in their future work context as educators.

For the development of this experience, we relied on the guidelines of the "ChatGPT and AI Use in Higher Education Guide" published by the United Nations Educational, Scientific and Cultural Organization [11]. These guidelines emphasize the importance of ensuring the accuracy of information when using it, guaranteeing human oversight, teacher mediation and community participation, training students on issues like academic integrity, the potential presence of biases in information, recognizing both the positive and negative uses of AI, and striving to maintain solid ethical principles grounded in the common good of society [11]. These themes were addressed and discussed with students before starting the learning experience, and also, at the end, through a group discussion on the advantages, challenges, and suggestions for the use of AI in their future role as teachers.

It is worth mentioning that this initiative was developed within the context of participating in an "AI Learning Community" convened by the Teacher Development Unit of the Vice-rectorate for Academic Affairs at the Pontifical Catholic University of Valparaiso, which involved teachers from different disciplines, aiming to promote reflection, analysis, and academic discussion on the possibilities and challenges of incorporating AI technologies in student training. In this context, practical experiences were also developed as a basis for progressively integrating AI into student training. The initial advances of this work were presented within the community to receive peer feedback, and adjustments were made to the implementation. Subsequently, the preliminary results of this work were presented through an audiovisual format at the "Innovation Week for Training" organized by the university in the second semester of 2023.

The article is organized as follows: it first contains a theoretical framework reviewing concepts and main theoretical discussions on the implications, risks, and ethical aspects affecting the use of AI in the university context and in training future educators. Additionally, the methodology section presents the design type, instruments, and participating subjects, along with the different phases of integrating the chatbot into the learning experience with pre-service teacher students. Then, the results are presented, organized based on the techniques and instruments applied, first synthesizing the results of the prior knowledge questionnaire KPSI, and then presenting the qualitative analysis of the questionnaire and Focus Group, through the establishment of emergent codes and their subsequent grouping into categories that explain the phenomenon under study, in this case, the students' perceptions. Finally, the study derives the theoretical and practical implications of the research, also acknowledging its limitations and proposing directions for future investigations.

Since its introduction in 2022, ChatGPT has gathered a great attention owing to its remarkable abilities in natural language processing. This large-scale language model, distinguished by its characteristics of conversational interface, faster responses, and costs-effectiveness, was an evolution from GPT-3 model [12]. This technology has demonstrated potential across diverse sectors including education, business, and computational social systems, showcasing its wide applicability [13, 14]. However, ChatGPT faces limitations, in areas such as narrative coherence, bias, and the dissemination of fake news, as reported by [15]. The development of chatbots such as ChatGPT is indicative of a broader trend in AI's role in transforming customer experiences and interactions across numerous industries [16]. [7] argue that ethical considerations, societal implications, and the design of these technologies must be addressed to ensure their reliable and beneficial integration into society. Continuous research and development concerning ChatGPT underscore its capacity to revolutionize the processing, dissemination, and application of information across various domains [14].

2.1 AI Challenges in Higher Education and Future Teacher Training

The challenges posed by Artificial Intelligence (AI) in higher education and pre-service teachers training are multifaceted. AI applications in education, such as personalised learning, automated grading, virtual tutoring, and learning analytics, offer significant benefits but also raise ethical concerns [1]. The shift to online teaching during the pandemic has underscored the importance of multiple platforms to enhance accessibility and support for online learning for pre-service teachers [17]. Moreover, preparing future teachers, particularly in terms of technological integration skills and competencies, plays a crucial role in addressing the challenges faced by novice teachers [10].

2.2 Digital literacy, data literacy, AI literacy, and teacher-mediated AI tools

The triad of digital literacy, data literacy, and AI literacy has been explored by [18], who demonstrated their similarities and intersections, while also highlighting the pivotal role of teachers as mediators of AI tools for their effective use in the classroom. Research conducted by [19] underscores the significance of developing strategies for the responsible implementation of ChatGPT in educational settings, aiming to cultivate informed discussions about its application. This underscores the essential need for teacher mediation to ensure that student engagement with ChatGPT is consistent with educational objectives and ethical standards. [20] addresses the growing importance of formal writing skills in academic and professional contexts, underlining the value of effective communication in the digital era. This accentuates the role of teachers in directing students on how to interact proficiently with AI technologies like ChatGPT to enrich their educational experiences. Furthermore, [21] investigates students' views on the efficacy of learning mathematics with ChatGPT, suggesting that it can assist students in concentrating on challenging concepts, thereby enhancing their learning outcomes.

2.3 Academic Integrity

ChatGPT has been placed at the centre of discussion because its potential impact on academic honesty and plagiarism. [22] studied this challenge and stated that universities must take proactive and ethical measures to the use of these tools. While some scholars have expressed concerns about issues such as misinformation generation, bias, and privacy associated with ChatGPT [23], others have highlighted its potential uses in knowledge assessment [4]. However, the use of ChatGPT in education raises important considerations for academic integrity and student learning [24]. As artificial intelligence applications like ChatGPT continue to advance, there is a growing need to address ethical challenges, transparency, and accountability in their implementation in educational settings [1, 25].

2.4 Implications for Assessment in Higher Education

In the domain of redefining assessment and feedback within higher education, the integration of ChatGPT and analogous AI technologies presents both opportunities and challenges. ChatGPT has been recognized for its potential to enhance personalized learning experiences, provide quicker and more regular feedback, and support knowledge assessment [24]. However, the emphasis on AI tools like ChatGPT, primarily as aids for improving task feedback and administrative tasks, rather than addressing concerns related to academic integrity, is notable [24]. This underscores the importance of carefully considering the broader implications of AI integration into educational assessment processes [26]. [8] stated ChatGPT be able to create new ways of authentic assessment in higher education.

3.1 Research method

The present study employed a sequential mixed-methods methodology with a qualitative-interpretative approach [27], which is most pertinent to the scope of the objectives. This approach allows for the analysis of participants' perceptions and experiences within a particular context or situation [28]. In this case, it aligns with the purpose of the study, which seeks to interpret and describe students' perceptions regarding the integration of ChatGPT into higher education as an assistant and the implications of AI use in their future teaching roles within a workplace context.

The methodology employed was that of a case study, designed to explore a phenomenon within a specific context or among a distinct group of students. In this case, the focus was on students from various teacher education programs who were enrolled in the 'Assessment for and of Learning' course during the second semester of 2023. These individuals, being in their fourth year and concurrently participating in an intermediate practicum at schools or high schools, formed a unique cohort. Their status as future educators, engaged in both theoretical studies and practical training, rendered this group particularly interesting. Their dual roles and experiences provided valuable insights into their perceptions of the use and impact of AI in educational settings, both from the standpoint of students in training and as prospective teachers.

This methodological approach facilitates the analysis and formulation of proposals grounded in empirical data, including participants' discourse as documented in textual responses or articulated during group discussions. Accordingly, a questionnaire featuring open-ended questions was administered, followed by a focus group discussion conducted at the conclusion of the experience. These methods aimed to capture students' perceptions regarding their experience with utilizing ChatGPT as an assistant in developing the workshop outlined in the course (see Fig. 1).

An open-ended questionnaire was developed, drawing upon central themes identified in previous studies, to explore students' perceptions during practical experiences involving the use of chatbots as assistants. To ascertain the content validity of the questionnaire, a thorough analysis of literature reviews and analogous prior research was undertaken. This analysis highlighted several key themes: students' understanding of chatbots, their prior interactions with these tools, and the various ways in which students utilize ChatGPT throughout their university education [3, 5, 29, 30]. These identified themes informed the design of the Prior Knowledge Questionnaire (KPSI), which subsequently underwent validation through expert review by one research specialist and another academic specializing in assessment. This evaluative process facilitated the refinement of certain statements for greater clarity and more precise formulation.

The design of this study was based by the principles and recommendations outlined in the 'Guidelines on the Use of Artificial Intelligence and ChatGPT' published by [11]. According to these guidelines, ChatGPT can fulfil various roles within the teaching and learning process, thereby augmenting the students' learning experiences [11]. Consequently, ChatGPT was employed in the planning and execution of this workshop, which is considered pivotal as it encompasses the course's learning objectives related to students' competencies in the assessment domain. Specifically, it facilitates the design of criteria, assessment scenarios, and tools that students will subsequently employ in practical interventions conducted during the semester.

3.2 Planning and Implementation Phases

3.2.1 First Phase

The experience was designed and introduced to the students in the first week, with an invitation for active participation. The objectives were clearly articulated, and the structure of each phase, along with the details of the workshop incorporating ChatGPT as an assistant, was thoroughly explained. This workshop was pivotal, aligning with a key learning outcome aimed at enhancing the course's competencies. It involved the development of evaluation criteria along with their respective quality levels or standards, forming the foundation for the creation of assessment scenarios and the design of evaluation rubrics. These elements were intended to be applied in the students' practicum schools within a real-world context, following a final review by the instructor.

It is important to highlight that, preceding this workshop, an in-class activity was conducted wherein students were instructed on the process of formulating evaluation criteria based on objectives and cognitive skills for an initial formative experience. Following this activity, students received feedback from both the teacher and the assistant. This feedback was provided in written form for each student's work via the course's virtual classroom and orally, summarizing key points for the entire course group. The feedback emphasized the strengths of the work and offered suggestions for improving the proposed evaluation criteria. Consequently, students entered the workshop with prior knowledge on criteria formulation. This foundational understanding was crucial for enabling them to propose quality evaluation criteria and effectively discern the information received from ChatGPT during the workshop.

Prior to initiating the workshop, the Knowledge and Prior Study Inventory (KPSI) questionnaire, first published in 1977 by Young & Tamir [31], was administered. Since its initial publication, the KPSI has undergone updates and has been extensively applied across various domains of knowledge and in student training. [32] revisited the instrument in a subsequent publication, highlighting its use in diverse projects and research endeavors spanning over 15 years. Within these studies, the KPSI's reliability and validity as a self-assessment tool have been affirmed. Furthermore, it serves as a means for educators to gather insights into students’ prior knowledge concerning specific topics or skills to be explored during a course.

This instrument enables teachers to collect valuable data regarding students' perceptions of their understanding prior to commencing a specific topic within the course. On one hand, it facilitates autonomous reflection among students on their prior knowledge; on the other hand, it provides faculty members with a foundation to tailor teaching strategies, thereby optimizing the pedagogical process or intervention. For the purposes of this study, the instrument was utilized at the outset to gather information about students' prior knowledge and any experiences they may have had with ChatGPT. This information was then used to adjust the design of the pedagogical process (see Table 1).

Table 1

Questions & Categories
Question nr.	Question statement	Categories
Q01	The characteristics and functions of ChatGPT as a virtual assistant for students in the university context	Knowledge
Q02	Distinguish different AI tools, with respect to their characteristics and functions	Knowledge
Q03	The use of ChatGPT, understanding its operation, and how I can use it as an assistant in my student training	Experience & Use
Q04	The relevance of formulating questions or prompts so that the chatbot works efficiently and meets the requirements	Experience & Use
Q05	The core characteristics of prompt formulation, such as: determining a clear purpose, a particular context or case, defining an audience, delimiting the topic, personalising or asking it to play a role, among others	Experience & Use
Q06	The search and information analysis strategies that ensure the reliability of the information provided by the chatbot or other AI tools	Experience & Use
Q07	The challenges of using information from ChatGPT in assignments because it can deliver imprecise or unrealistic data (hallucinations)	Student perception
Q08	The risks regarding the use of data and information from people interacting with chatbots	Student perception
Q09	The importance of ethical principles and academic integrity declared by the university for handling information and using any AI tool	Student perception
Q10	The challenges that AI tools pose in teacher mediation and in our future teaching role	Student perception

Students were requested to assess their level of prior knowledge on specific academic contents using a scale of varying complexity, which provided four levels of self-evaluation. The questionnaire, designed for this study, consisted of 10 statements accompanied by a 1 to 4 scale. Here, level 1 represents minimal prior knowledge of the topic under discussion, while level 4 indicates the most extensive knowledge and experience with the chatbot. These statements were categorized into three groups based on a preliminary analysis of literature and previous research on the use of chatbots in higher education settings. The categories included: understanding of AI, experiences and uses of AI tools, and modes of use and perceptions concerning AI. The collection of data on students was confined to identifying the teaching career or specialty they were pursuing. This limitation was intentional, as the aim at this phase was to explore the prior knowledge of all students, irrespective of their academic background.

In this initial phase, several class sessions were dedicated to analysing and discussing the implications of utilizing ChatGPT. These discussions covered the risks associated with personal data interaction, potential biases in the information provided, and the significance of maintaining academic integrity when employing AI tools. This effort reinforces the university's ongoing commitment to transversally educating students about this critical ethical dimension across all subjects. To delve into these issues, a segment from the UNESCO guide on 'Challenges and Ethical Implications' was distributed to students for the analysis of key topics: academic integrity, data protection, cognitive biases, and equity in accessibility [11].

3.2.2 Second Phase

In the second phase, students received a comprehensive explanation of the workshop during a class session, which was subsequently published in the course's virtual classroom. In this workshop, students were guided through instructions and a predefined format to develop evaluation criteria that align with the curriculum of their specialty, as well as the objectives and competencies outlined for the level or course in which they will undertake their practicum. Utilizing these criteria, students were tasked with designing an evaluation scenario and their related instruments. Additionally, they were briefed on the process for formative submissions of their work. This involves submitting an initial draft of their progress via the virtual classroom to receive feedback, followed by a final, enhanced submission. Upon completion, students are prepared to implement the evaluation scenario in the schools where their practicum is conducted. Notably, the criteria for assessing their work were also communicated to the students during these sessions.

In the process of conducting this workshop, which spans approximately one month with two class sessions per week, the chatbot was utilized as an assistant to support various activities. These activities included refining the formulation of evaluative criteria proposed by students, providing theoretical information about these criteria while contextualizing them within the discipline, and offering feedback on their proposed evaluation scenarios. Additionally, students were encouraged to independently research and experiment with other potential chatbot applications, always within the context of the workshop's activities. This pedagogical strategy aims not only to enhance the students' learning experience through the use of advanced educational technologies but also to cultivate a proactive and exploratory approach to learning, qualities that are paramount in higher education. Furthermore, students were instructed to maintain a journal documenting their interactions with the chatbot to facilitate future evaluations of the experience.

3.2.3 Third Phase

To conclude, after the students completed their workshops and submitted them to the virtual classroom for feedback, a brief questionnaire (Appendix, Open-ended Questionnaire) with open-ended questions was administered to gather insights into the students' perceptions and experiences with using the chatbot as an assistant in an authentic learning scenario. The questionnaire was administered in-person to ensure the immediacy of students' responses during a class session (see Table 2).

Table 2

Overview of the open-ended survey categories and its associated descriptions
Category	Description of the information that is collected or requested
Use and experiences of students during the implementation of ChatGPT as an assistant	Participants were asked to explain how they had used ChatGPT during the implementation, its usefulness and limitations, justifying their assessments.
Possibilities for the use of ChatGPT during their training	Different applications that participants explored independently with ChatGPT within the workshop context.
Students' perceptions of the implications of ChatGPT in the educational context from their role as pre-service teachers	A discussion is presented with quotes extracted from an opinion column in which a series of questions about the use of ChatGPT in schools are raised, and students are asked to analyse, discuss, and present a well-founded point of view on the issues presented.

The development of the questionnaire was informed by a review of the literature [3, 5, 29, 30] with a focus on the criteria used to explore students' perceptions of AI, specifically their experiences with using chatbots as an assistant during the workshop in the course. To establish the content validity of the questionnaire, a matrix was created based on the objectives and criteria derived from the literature review. This was followed by a validation process through expert judgment, which assessed the questions for clarity, relevance, and appropriateness, resulting in minor suggestions for adjustments in their phrasing or wording.

Once the questionnaire implementation was concluded, a script for a Focus Group was developed (Appendix, Focus Group Protocol and Script), incorporating two of the key themes addressed in the previously used instrument. This Focus Group also aimed to promote participation and critical reflection among students on the various topics discussed during the experience and on their future professional performance as teachers with the advent of AI in the educational context. For this purpose, a segment was introduced in the focus group where a text with quotes was presented, and a discussion about the impact of artificial intelligence on student learning processes in secondary schools was initiated. Various questions were raised about the role of the teacher as a facilitator or mediator in these educational processes.

For the implementation of the Focus Group, students were invited to participate in the final class session of the semester. This invitation was extended with the assurance of reliable and anonymous handling of data, encompassing both the recording and transcription processes. Participants were selected from pairs who had engaged in the experience and completed their practical immersion at a practicum school during the semester. The Focus Group, comprising 12 students, was conducted with their consent, and was duly recorded. The objectives and the expected forms of interaction and participation were reiterated at the beginning of the session.

To ensure the integrity of the process and bolster student confidence, the assistance and mediation of a teaching assistant from an advanced course were sought. This assistant, affiliated with one of the disciplines offering the subject that semester and nearing the completion of their studies, had prior exposure to courses related to educational research. Therefore, they were well-positioned to make a valuable contribution to this process.

3.3 Sample

We collected KPSI questionnaire data from 26 pre-service teachers, whereas 24 participated in the open-ended questionnaire. This subset included 12 pairs of pre-service teachers who completed their practicum in cooperating schools. Due to size restrictions imposed by the methodology, one student from each pair participated in the final focus group session.

The analysis of the results was conducted with rigor to align with the research objective and maintain coherence with the central themes outlined in the theoretical framework regarding university students' perceptions of AI use. Validation of the findings was achieved through a data triangulation process, incorporating the various techniques employed. The organization of the analysis and the presentation of the main findings were structured according to the phases of the research. As such, the descriptive statistical analysis of the Knowledge and Prior Study Inventory (KPSI) will be presented initially, followed by an overview of the didactic changes and adjustments made to the design and subsequent intervention. The analysis of qualitative data and the outcomes of the questionnaire administered in the second phase will then be discussed, highlighting students' perceptions of utilizing the chatbot as an assistant in their learning process, as derived from the experience conducted. Finally, the qualitative findings from the Focus Group will be revealed.

4.1 Prior Knowledge and Student Experience in Using ChatGPT (KPSI)

An analysis of the data combined the values obtained from the KPSI and the categories defined for each of the questions. A descriptive statistical analysis was then conducted with the following results:

4.1.1 Knowledge Category

Questions:

Q01: Characteristics and functions of ChatGPT as a virtual assistant for students in the university context.

Q02: Distinguishing different AI tools, regarding their characteristics and functions.

Insights: Focusing on the fundamental knowledge of AI tools, the moderate level of agreement (average response approximately 3.02) suggests that respondents have a relatively solid understanding of the characteristics and functions of AI tools. This area appears to be a strength among respondents, though there is still room to deepen this knowledge, particularly in understanding various AI tools and their specific applications in educational contexts.

4.1.2 Experience and Use Category

Questions:

Q03: Using ChatGPT, understanding its operation, and how I can use it as an assistant in my student training.

Q04: The relevance of clearly formulating questions or instructions when interacting with AI tools like ChatGPT.

Q05: Fundamental characteristics of formulating instructions for effective interaction with AI tools.

Q06: Search and information analysis strategies that ensure the reliability of the information provided by the chatbot or other AI tools.

Insights: This category, which had the lowest average response (approximately 2.66), underscores a significant gap between theoretical knowledge and practical application. It highlights a critical area for improvement, showing that, although respondents may understand AI concepts theoretically, they feel less confident in their practical application. This suggests the importance of integrating more practical experiences and real-world scenarios into learning pathways to enhance practical skills and confidence in the effective use of AI tools.

4.1.3 Student Perception Category

Questions:

Q07: The importance of ethical principles and academic integrity stated by the university for information handling and the use of any AI tool.

Q08: Challenges that AI tools pose in faculty mediation and our future role as teachers.

Q09: Perceptions of the impact of AI on the future of education and the teaching profession.

Q10: Student views on the integration of AI tools within educational environments and their potential to enhance learning.

Insights: Maintaining the highest average response (approximately 3.37), this category reflects a strong agreement or positive perception regarding ethical considerations, challenges, and the potential of AI in education. This suggests that respondents are particularly aware and optimistic about the broader implications of AI, its ethical integration, and its transformative potential in educational settings.

The analysis highlights the disparity between theoretical knowledge and practical application among respondents. It emphasizes the need for educational strategies that not only enhance understanding of AI tools but also focus on developing practical skills and confidence in applying these tools in real-world contexts. This approach will better prepare students to effectively leverage AI technology in their future academic and professional lives, ensuring they can navigate the evolving landscape of AI in education.

During the same week, based on the results, two types of activities were conducted in a session: writing prompts, as several students were unfamiliar with this topic. Therefore, suggestions for drafting them were presented, and examples were illustrated in a presentation. Subsequently, students practiced different assignments by asking the chatbot to propose evaluation criteria for a specific assessment situation or to contribute feedback to the criteria they had previously developed formatively or as a proposal.

4.2 Qualitative Data Analysis

4.2.1 Survey Analysis

We present the results in alignment with the study's objective, which aimed to interpret and describe students' perceptions of utilizing ChatGPT as an assistant within a contextualized learning experience during their training, as well as the implications of this AI tool in their prospective work contexts as teachers.

To analyse the data, responses from the open-ended questions in the questionnaire and the Focus Group were transcribed. This facilitated the processing of analysis units and the establishment of codes with the assistance of Atlas.ti software.

The analysis employed qualitative content analysis methods. Following the steps of this methodology, emergent codes were initially identified from the students' discourse, both written and oral. This process began with the questionnaire data, followed by the Focus Group data. To enhance the rigor and validity of the analysis, the initial coding was conducted independently by both the lead researcher and a research collaborator. A subsequent meeting was arranged to review and discuss both sets of codifications, refining the coding process to consolidate the codes.

The interpretive process, grounded in the established codes, aimed to identify patterns, themes, and connections among the codes, which were then organized into central categories. Categories were first identified for the questionnaire data and subsequently for the Focus Group data. Given the study's scope, codes that were descriptive and explanatory of the phenomenon under investigation were selected. This approach aimed to capture the students' perceptions, reflecting the richness of their dual perspective: as trainees and as future educators, navigating their practicum experiences.

Initially, we will discuss the main findings and categories identified in the questionnaire, concentrating on the students' perceptions as university students within the context of their experiences using the chatbot in the course. The following are the analysis categories and their corresponding codes.

Students perceived positive aspects of interacting with the chatbot as its assistance and guidance to co-construct evaluation criteria with their respective quality levels and/or propose situations or forms of assessment. Most students stated that it had been useful to refine, delimit, or establish quality standards for the criteria they had previously set or had as a preliminary proposal. In this context, they also used it to support their own ideas or proposals for assessment situations for the courses they were working with in the practicum schools, giving the chat a personalized role and clear and specific information about the context, objectives, and evaluation criteria they were working on. In this sense, they expressed the relevance of using strategies to draft prompts with as much contextual information as possible and clearly directing the type of response or information being requested (see Fig. 2).

Some participants stated the following:

We first focused on how ChatGPT operates, the more detailed and more information we provide, the better its response will be. With this in mind, we gave it parameters and guided it by giving it an identity, for example: You are a teacher of children aged 12 to 13 years old, working in a private school in Chile and teaching music, aiming for students to be able to perform some musical pieces... You have extensive experience working with neurodivergent students, based on this, what would you consider when developing an assessment situation? (questionnaire/pair 3).

We used it to provide conceptual support to our criteria, definitions, or delimitations from different authors or music specialists, to help us set the quality of the criteria and the levels or standards, because evaluating 'musical performance' in primary is not the same as in a higher course" and it gave us a definition from different authors that served our work, after that we defined three to four criteria that we used to develop a rubric. (questionnaire/pair 1).

We asked for references to better formulate the criteria we had already created based on a taxonomy of skills centred on mathematics, to assess the use of the drafting of notable products and then their development in algebraic expressions. It provided us with a list of verbs associated with skills or aspects to evaluate, with that we reinforced the evaluation criteria and then asked it to propose situations in which the notable products must be applied concretely or in everyday life, for an assessment of students. (questionnaire/pair 2).

Another function highlighted by the students was the guidance and feedback the chatbot provided on their work, having a constant assistant during the workshop to whom they could ask questions, request to review the progress of their work, and ask for quick feedback contributed to the development of their work in its first phase, as after the formative submission, they always have a feedback space with the assistant and teacher. For example: "Sometimes you need an external viewpoint when you're doing work that's neutral, ChatGPT can be that and support or criticize or suggest improvements to the work". (questionnaire/pair 6).

However, at this point, it is worth noting that while the AI's support or feedback to improve student work was valued, they also stated that it could not replace the guidance and feedback of the teacher, as being a specialist in the subject and due to their expertise, they can make more focused comments on some aspects of the student's work, which for the chatbot might go unnoticed:

It served us to have an external look at the criteria and the assessment situation we had proposed, and we asked it to provide feedback on our work, even to improve more formal things like the clarity of the evaluation instructions, and with that, we saved time, we focused on adjusting the quality levels of our rubric which is the most difficult, but when the teacher and assistant gave feedback on the rubric, they were more demanding and focused on the differences between the various performance levels, that's why we believe, it cannot replace the feedback of the teacher as a specialist in their area. (questionnaire/pair 4)

The main challenge for students when using AI tools as assistance for their work is to always keep academic integrity in mind, and the rules regarding plagiarism or other aspects of authorship that govern university activities. For this reason, they stated that these tools should always be used as support and accompaniment, but not to replace the students' work during their university training (see Fig. 3).

Students also expressed the risk of potential inaccuracy and unreliability of the information provided by the chatbot, as the data could be imprecise or result from hallucinations, which is widely documented for this type of tools. However, this issue was analysed before the experience, so the students knew it was necessary to review and contrast the information delivered by ChatGPT. Regarding the context, they considered that the chatbot might issue responses that are not relevant to the conditions and characteristics of the student group at the practicum school where they must apply the assessment situation:

Not knowing how to classify, critique, or confirm the information provided by the chatbots, students can end up with imprecise information or with non-existent sources or poorly made references. (questionnaire/pair 5).

Using inaccurate or biased content thrown by ChatGPT or outdated information since the free version does not have internet access to update information. (questionnaire/pair 7).

We believe it is a very useful tool, but it depends on how it is used because it is still easy for students to copy things verbatim and fall into plagiarism. At least in our discipline, the fear of these tools is not something new; there was a time against calculators, now against apps that solve exercises connected with AI. (questionnaire/pair 5).

4.2.2 Focus Group Analysis

In this second part, we present the categories extracted from the analysis of the Focus Group transcript. This technique aimed to promote critical reflection among students and gather information on some relevant topics presented in the questionnaire that require more critical analysis. Group discussion mediated can promote dialogue and problematization on new topics or emerging scenarios, such as the advent of AI in the educational field. In line with the purpose of this study, the discussion focused on the students' perceptions from their future teaching role, considering that they have had practical interventions during the semester that could influence their perspectives.

From the analysis, 2 central categories were raised with their respective codes:

Students, or future educators, highlighted the pivotal role of teachers as mediators in training and guiding students, which is crucial for cultivating research skills and the analytical evaluation of information provided by tools like ChatGPT. This involves instructing students on distinguishing between reliable sources and data that can be used responsibly. The educational process should promote critical thinking skills, including analysis, synthesis, and the problematization of various topics discussed in school, and foster environments conducive to critical reflection on AI implications and the strategic use of knowledge to enhance skills that enable students to innovate or create based on the critical application of foundational knowledge (see Fig. 4)

Furthermore, future teachers underscored the importance of analysing and discussing with students the utilization of fake news to sway public opinion on diverse social and political matters. This ensures that young individuals are cognizant of the methods of circulation, production, and usage of information within new technologies.

One of the factors that could affect the mediation a teacher can provide to educate students from a critical standpoint on the use of these tools is the inequality or lack of equitable access, leading to a digital divide. This divide is exemplified by the absence of internet access in some schools or the lack of connectivity in many students' homes. Such circumstances can restrict their opportunities to explore and utilize AI technologies, undermining the principle of equity in education by preventing all students from equally learning about technologies that are shaping the future of society and the work environment.

That's why I think teachers first must educate students to be aware of the benefits and risks of AI. Also, prepare them to be critical of the information they find online, check their sources, for example, we all know that fake news exists and influences public opinion, and perhaps a debate on that with students could motivate them to critically reflect. (E2/Focus Group).

I think it's important that high schools or schools, teachers, and administrative teams work together to create some ethical framework or guidelines for the use of AI and share them with students and teachers explain their importance so that they can have responsible use. (E3/Focus Group).

Students voiced their apprehensions about future educational landscapes in which they will serve as teachers, grappling with the swift progression of artificial intelligence. They were already encountering these challenges during their practicum experiences in the schools where they implemented interventions that semester. In certain schools, the use of AI was banned, while in others, discussions were only just beginning, particularly concerning how to manage assessments such as research papers or other tasks assigned for completion at home without direct school oversight (see Fig. 5). The majority of students participating in the group discussion concurred that prohibition was not the optimal approach to this issue; rather, the strategies should focus, firstly, on educating and training students, and secondly, on revising assessment methods. In this context, one student specifically remarked:

We must transform the way of assessing; we cannot continue assessing in a reproductive manner or in mathematics with a set of routine exercises or problems. We need to design assessments that promote mathematical thinking and reasoning in different situations, which are close to the students' everyday problems. (E8/Focus Group).

Another student on the same topic stated:

The teacher must give meaning to the assignments or tasks, in which students have to analyse, solve problems, contrast different viewpoints, be more critical, and there the use of chat or other tools can be enhanced, so students benefit from ChatGPT as a support in the foundation of ideas or responses. (E9/Focus Group).

Concerning the professional future of students there exists an escalating uncertainty among pedagogy students about their career prospects. This uncertainty stems from the fact that, although AI technologies can facilitate novel teaching and learning methodologies, they concurrently pose significant queries regarding the teacher's role and their adaptability to an ever-evolving professional landscape. Within the group discussion, students deliberated and examined the function of the school as a community in formulating strategies and attitudes towards working with AI, highlighting the critical importance of teacher training in employing these new tools for educational purposes. On this matter, one student suggested:

We grew up in a constantly changing society where technology is increasingly present, and we have been adapting to it, so, in this case, with artificial intelligence, we would have to do the same, look for ways to work with it in the educational context with our students. (E9/Focus group).

I think it's important that high schools or schools, teachers, and administrative teams work together to create some ethical framework or guidelines for the use of AI and share them with students and explain their importance for the responsible use of these tools. (E10/Focus Group).

To conclude the analysis, students highlighted the social role of the school as a space where students not only develop their cognitive but also social skills. The school is a privileged space to learn to live together in a democratic society and interact with others who think differently and come from different spaces, traditions, or cultures. For students, one of the central purposes of the school is to form well-rounded individuals, considering all their dimensions, who can collaborate and contribute to the society they are part of.

The fear of AI arises from various factors, as it can replace many tasks currently performed by humans, creating uncertainty:

At first, I was worried, but then I thought that the school has many more functions that are very important, such as comprehensive training and the social interaction of a person, which cannot be replaced by AI. (student 10/Focus Group).

In this section, we discuss the main findings that emerged from the analysis of results regarding the central question of this research, which focused on how pedagogy students in a fourth-year evaluation course related to practicum perceive the integration of ChatGPT as an assistant in a learning experience and their perceptions of the implications of this tool in their future professional teaching performance. Based on the main findings and the coding process carried out, we highlight the following dimensions: students' perceptions of the integration of AI as an assistant during their training and students' perceptions of the implications of AI in their future professional teaching role in the educational context. Within each of these, emerging themes from the central categories will be discussed in contrast with findings from other research presented in the literature.

5.1 Students' perception of the use of ChatGPT as an assistant during a learning experience in assessment assignments

Students perceived positive aspects of interacting with ChatGPT as an assistant during the workshop they conducted, specifically valuing its contribution as a co-designer of evaluation criteria, providing theoretical or conceptual support, defining the quality or standard of criteria, or giving feedback on criteria they had preliminarily outlined or proposed as ideas. They also mentioned the usefulness of the chatbot in providing ideas or proposals for evaluations based on objectives and criteria, specifying the context of application and the characteristics of the students. For the students, having an assistant that could offer quick assistance, answer their questions, and provide feedback at any moment or phase of their work development was considered valuable. This positive point of using this tool converges with the findings of other research on university teaching applications where students value having constant and personalized support [30, 33, 34].

Furthermore, students mentioned the chatbot's usefulness in providing ideas or proposals for evaluations based on objectives and criteria, specifying the application context and student characteristics, sometimes seeking ideas for diversifying evaluations for neurodivergent students. At this point, students highlighted the importance of developing strategies for writing prompts in a personalized way, with clear requirements and contextual information. Other studies have also suggested that when appropriate inputs are given to the chatbot, it can contribute structures for evaluations, question ideas, and feedback [24]. It's worth noting that in the first phase of implementing this experience with students, guidelines and examples were provided on how to interact with ChatGPT. As [5] points out, it is very relevant for teachers to guide students on techniques for giving instructions and moderating dialogues established with the chatbot.

This prior preparation of students on the use, writing of prompts, and form of interaction with ChatGPT was based on preliminary findings about students' prior knowledge and experiences. Like in other studies, although most students scored high on knowledge of AI tools, they had low scores regarding practical handling or mastery of its operation and forms of interaction, with only a limited number of students informed about the importance of managing interactions to use AI tools effectively [3, 5]. A notable gap was observed between theoretical knowledge and practical application, emphasizing the need for more practical experiences to improve students' confidence in the effective use of AI tools.

Regarding the risks or challenges presented by the use of this AI technology in the higher education context, students recognized the limitations of the information provided by ChatGPT, in terms of its validity, trustworthiness, and specificity to respond to a particular context. Information credibility issues have been reported by other research; in a study conducted by [5] with engineering students, while students valued the benefits of ChatGPT, they were cautious about the possibility of inaccuracies in its results. To minimize these risks, it's important for students to have knowledge about the topics or procedures they are working with the AI assistant, as this will allow them to discriminate and analyse the quality of information or responses arising from the interaction, contrasting the information provided [5,35].

However, they also recognized that the tool could not replace the guidance and feedback of a teacher, especially in providing focused comments on specific aspects of their work [36]. Additionally, the analysis made evident that while students valued the support and feedback provided by ChatGPT in improving their work, they also recognized the limitations of AI tools in providing more generic feedback with quality levels different from those requested in the course or in their practicum processes; therefore, feedback based on the teacher's expert knowledge could not be replaced, especially when providing comments focused on specific aspects of their work [36].

Lastly, students expressed that it was important, especially for the new generations of university students who are entering with the use of AI, to keep in mind academic integrity and the existing regulations at the university about plagiarism and responsible use of these tools. In the

context of this research, the university had been working since the time of the pandemic and through various strategies on plagiarism prevention. In this line, it is recommended to address this issue in an institutional and cross-cutting manner across different areas of the university. Suggestions from other studies, such as those by [22], also emphasize the importance of educating students about the implications of plagiarism, working with formative assessments that detect it in time, and suggesting changes or improvements to students.

5.2 Students' perceptions of the implications of ChatGPT in their future role as teachers

Focus Group discussions highlighted the essential role of teachers as trainers and mediators in the use of AI technologies with adolescents or young people who are in a developmental process and therefore more exposed to indiscriminate use of technology. In this case, future teachers raised as risks to consider the possibility that the critical thinking skills of young students could be affected by the use or dependence on AI technologies. Against this, the importance as teachers of using teaching and learning strategies that contribute to the development of critical thinking while progressively integrating artificial intelligence tools.

Critical thinking refers to the thought process that involves various skills such as interpretation, analysis, inference, self-regulation, and evaluation for the formulation of judgments that will affect viewpoints, beliefs, and decision-making on different relevant social issues [37]. This skill is central to 21st-century education and can be progressively developed throughout the entire educational trajectory from secondary education and throughout life, allowing people to participate critically in a democratic society [29]. At this point, students suggested that teachers could contribute to the training of young people in the analysis and critical evaluation of information disseminated through social networks and that in some cases may be created by AI tools, for example, they could use examples of how to unveil fake news, and discuss with students how through these communication strategies public opinion can be generated and influence the political debate of a democratic society. In this line, there are research studies that show experiences where AI and information technology are used to promote critical thinking, but they are not conclusive as to the direct contribution to its development, rather the evidence could suggest that the implementation of innovative technologies, like chatbot systems, could motivate students to reflect and analyse the information they receive, thus facilitating skills linked to the development of critical thinking [29, 38].

Regarding the risk factors associated with the use of these tools in secondary education, such as the possible lack of reliability in the information provided, the need for its analysis and corroboration, along with the analysis skills that young people must develop to interact responsibly with AI, the findings revealed that one of the strategies to mitigate these risks is the training of students and how the teacher guides students to develop these skills while integrating or working with AI. Other studies have also coincided in the importance of having a tutor or with the mediation of the teacher so that students can use AI ethically, responsibly, and carrying out the processes of review and analysis of their answers or information [10, 30, 39].

They also analysed the impact of AI in education, underlining the importance of teacher training in leveraging these tools for educational purposes and establishing ethical guidelines for their responsible use within educational institutions, beyond merely prohibiting their use. Future teachers expressed the need to adjust teaching and assessment strategies to effectively use AI tools [36]. The need to transform or change assessment methods was one of the most relevant and important points, since the reflection and analysis of pedagogical practices in assessment in this subject is one of the learning outcomes expected to be achieved, and through this implementation, students were able to question more traditional practices and become aware of creating assessment instances that promote problem-solving, critical thinking, creativity, among others. The challenge presented by new AI technologies to assessment methods is also convergent with the findings of other research, in which methods associated with authentic assessment are proposed through solving relevant social issues, working with simulations and/or working on projects in subjects or interdisciplinary [3, 22, 39]. Regarding the assessments of the disciplinary areas of the participating students, future mathematics teachers expressed the importance of reviewing mathematics assessments that are based solely on operations and calculations of problems or tasks that refer to skills, such as calculating, since tasks are solved through the direct and immediate application of some algorithm or technique. The proposal is to move towards assessments in which mathematical thinking skills associated with the national curriculum are worked on, such as: solving problems, modeling and representing, arguing and communicating solutions based on procedures and mathematical reasoning, integrating, when relevant to the objectives, digital applications or tools that support or assist in problem-solving. In different national and international research studies, assessment in mathematics has been discussed as an instance that is part of learning,

with tasks or activities that demand higher reflection, dealing with every day, but complex topics that integrate various skills and present relevant problems to solve [40, 41].

In summary, the study emphasizes the need to integrate more practical experiences into educational strategies to better prepare pedagogy students for their future professional performance in the face of new challenges presented by AI technologies in the educational context, considering that the development of their digital skills is enhanced in applied experiences with the use of technology in environments where they have the guidance and orientation of teachers. These experiences will undoubtedly impact their future teaching role, since, as [18] suggest, they play a dual role, on one hand, teachers in training are also in a process of formation, and on the other hand, they are multipliers as teachers in schools by deploying their competencies for the training of future generations.

In this research, we have studied the perceptions of a group of pedagogy students on AI assistance during their training and the implications this new technology will have on their future performance as teachers in educational institutions. To develop our implementation, we were guided by the orientations and guidelines issued by [11] on training students in ethical aspects, academic integrity, critical analysis of the information provided by AI, and ways to interact with tools so that they contribute or assist in learning processes. Based on the study's findings, it was determined that the role of educators is crucial in training, guiding, and directing students in the responsible, ethical use of AI tools and promoting skills of analysis, discrimination, and evaluation of information, linked to the progressive development of critical thinking. The findings suggest that a balanced approach that combines theoretical understanding of AI technologies' operation with practical experience is central to preparing students for the professional challenges they will face as teachers in educational contexts with constant advances and changes from new technologies. This emphasizes the need to integrate more practical experiences into educational strategies to better prepare students to leverage AI technology in their professional field. For future teachers to transfer the experiences lived in the use of AI during their training and implement these teaching strategies in their pedagogical practices in schools, they require solid educational knowledge and the development of digital competencies. This is especially important in teacher training to make AI accessible and practical in school environments [42]. 21st-century university students are part of a generation that has grown up in an environment marked by the rapid advance of computer technology and have experienced the changes and evolution of new technologies [2]. Furthermore, future teachers are digital natives who use technology in their daily lives. These experiences and living in a constantly changing society, according to the findings of this research, give students more security and confidence in their abilities to integrate the advancement of AI into their future pedagogical practices. But despite the experiences lived during the pandemic and the students' confidence, it is necessary to strengthen their training, as gaps were observed between theoretical knowledge about the tools and their concrete application in an effective experience, as previously stated. In this way, it is sought that future teachers develop the fundamental and basic cognitive skills that teachers require to interpret content generated by chatbots, as proposed by [10]. The research indicates that students view AI tools positively in tasks such as co-constructing evaluation criteria and providing guidance and feedback on their work. However, students also recognize the importance of maintaining academic integrity and being cautious of the potential inaccuracies and unreliability of the information provided by AI tools like ChatGPT. This underscores the need for educators to instruct students on responsible use and critical evaluation of information obtained through AI tools. Finally, students recognize the need to address issues such as access barriers to technology in secondary schools, especially in more vulnerable schools and those far from urban areas, where some of the students conduct their practical interventions. This situation can create a digital divide and lack of equity in access to new AI technologies, aspects that must be considered to prevent disparities in learning opportunities. This characteristic of the educational context in some regions of South America must be considered in public and educational policies that promote the use of new technologies while safeguarding inclusion in educational processes.

Code Availability

The Python code used to process the KPSI survey in this study is available upon request.

Compliance with Ethical Standards

This study involving human participants was conducted in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments. Verbal informed consent was obtained from all individual participants included in the study. Given the non-invasive nature of the research and the absence of personal or sensitive data, our institution (Pontificial Catholic University of Valparaiso) does not require formal ethics approval for such studies. We confirm that all necessary steps were taken to guarantee the anonymity and confidentiality of the participants.

Informed Consent

Verbal informed consent was obtained prior to the KPSI survey and focus group activities.

Funding Not applicable

Conflicts of interest/Competing interests Not applicable

Author Contribution

M.A.P. and D.F.B. wrote the main manuscript and D.F.B. prepared tables & figures. M.A.P. developed the conceptualization, methodology, data gathering and analysis.D.F.B. elaborated the data analysis, tools selection and Python code writing. All authors edited and reviewed the manuscript.

Data Availability

The data that support the findings of this study are available from the PUCV Pedagogy School, but restrictions apply to the availability of these data, which were used for the current study and are not publicly available. The data are, however, available from the authors, who can request the corresponding permission from the PUCV Pedagogy School upon reasonable request.

Akgün S, Greenhow C. Artificial Intelligence in Education: Addressing Ethical Challenges in K-12 Settings. AI Ethics. 2021; https://doi.org/10.1007/s43681-021-00096-7
Zhao Y, Pinto Llorente AM, Sánchez Gómez MC. Digital competence in higher education research: A systematic literature review. Comput Educ. 2021; https://doi.org/10.1016/j.compedu.2021.104212
Kelly A, Sullivan M, Strampel K. Generative artificial intelligence: University student awareness, experience, and confidence in use across disciplines. J Univ Teach Learn Pract. 2023; https://doi.org/10.53761/1.20.6.12
Rudolph J, Tan S, Tan S. ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? J Appl Learn Teach. 2023; https://doi.org/10.37074/jalt.2023.6.1.9
Shoufan A. Exploring Students’ Perceptions of ChatGPT: Thematic Analysis and Follow-Up Survey. IEEE Access. 2023; https://doi.org/10.1109/ACCESS.2023.3268224
Hwang G-J, Chang C-Y. A review of opportunities and challenges of chatbots in education. Interact Learn Environ. 2023; https://doi.org/10.1080/10494820.2021.1952615
Gordijn B, Have HT. ChatGPT: evolution or revolution? Med Health Care Philos. 2023; https://doi.org/10.1007/s11019-023-10136-0
Crawford J, Cowling M, Allen K-A, Monash University, Australia. Leadership is needed for ethical ChatGPT: Character, assessment, and learning using artificial intelligence (AI). J Univ Teach Learn Pract. 2023; https://doi.org/10.53761/1.20.3.02
Tunjera N, Chigona A. Investigating Effective Ways to Use Artificial Intelligence in Teacher Education. Eur Conf E-Learn. 2023; https://doi.org/10.34190/ecel.22.1.1625
Kerneža M. Fundamental and basic cognitive skills required for teachers to effectively use chatbots in education. Sci Technol Educ NEW Dev Innov. Scientia Socialis Press; 2023; https://doi.org/10.33225/BalticSTE/2023.99
UNESCO. ChatGPT e inteligencia artificial en la educación superior. UNESCO; 2023.
Gupta B, Mufti T, Sohail SS, Madsen DØ. ChatGPT: A brief narrative review. Cogent Bus Manag. 2023; https://doi.org/10.1080/23311975.2023.2275851
Dwivedi YK, Kshetri N, Hughes L, Slade EL, Jeyaraj A, Kar AK, et al. Opinion Paper: “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy. Int J Inf Manag. 2023; https://doi.org/10.1016/j.ijinfomgt.2023.102642
Wang F-Y, Li J, Qin R, Zhu J, Mo H, Hu B. ChatGPT for Computational Social Systems: From Conversational Applications to Human-Oriented Operating Systems. IEEE Trans Comput Soc Syst. 2023; https://doi.org/10.1109/tcss.2023.3252679
Ray PP. ChatGPT: A comprehensive review on background, applications, key challenges, bias, ethics, limitations and future scope. Internet Things Cyber-Phys Syst. 2023; https://doi.org/10.1016/j.iotcps.2023.04.003
Nirala KK, Singh N, Purani VS. A Survey on Providing Customer and Public Administration Based Services Using AI: Chatbot. Multimed Tools Appl. 2022; https://doi.org/10.1007/s11042-021-11458-y
Tunjera N, Chigona A. Improve Pre-Service Teachers’ Online Learning Attendance and Accessibility Through Multiple Platforms. J Educ. 2022; https://doi.org/10.17159/2520-9868/i88a02
Kreinsen M, Schulz S. Towards the Triad of Digital Literacy, Data Literacy and AI Literacy in Teacher Education – A Discussion in Light of the Accessibility of Novel Generative AI. EdArXiv; 2023. https://doi.org/10.35542/osf.io/xguzk
Halaweh M. ChatGPT in education: Strategies for responsible implementation. Contemp Educ Technol. 2023; https://doi.org/10.30935/cedtech/13036
Fontenelle-Tereshchuk D. Academic writing and ChatGPT: Students transitioning into college in the shadow of the COVID-19 pandemic. Discov Educ. 2024; https://doi.org/10.1007/s44217-023-00076-5
Auna HSA, Hamzah N. STUDI PERSPEKTIF SISWA TERHADAP EFEKTIVITAS PEMBELAJARAN MATEMATIKA DENGAN PENERAPAN CHATGPT. HINEF J Rumpun Ilmu Pendidik. 2024; http://doi.org/10.37792/hinef.v3i1.1160
Cotton DRE, Cotton PA, Shipway JR. Chatting and cheating: Ensuring academic integrity in the era of ChatGPT. Innov Educ Teach Int. 2023; https://doi.org/10.1080/14703297.2023.2190148
Firat M. What ChatGPT means for universities: Perceptions of scholars and students. J Appl Learn Teach. 2023; https://doi.org/10.37074/jalt.2023.6.1.22
Sullivan M, Kelly A, McLaughlan P. ChatGPT in higher education: Considerations for academic integrity and student learning. J Appl Learn Teach. 2023; https://doi.org/10.37074/jalt.2023.6.1.17
Holmes W, Porayska-Pomsta K, Holstein K, Sutherland E, Baker T, Shum SB, et al. Ethics of AI in Education: Towards a Community-Wide Framework. Int J Artif Intell Educ. 2022; https://doi.org/10.1007/s40593-021-00239-1
Swiecki Z, Khosravi H, Chen G, Martinez-Maldonado R, Lodge JM, Milligan S, et al. Assessment in the age of artificial intelligence. Comput Educ Artif Intell. 2022; https://doi.org/10.1016/j.caeai.2022.100075
Bamberger M. Introduction to methods in impact evaluation. The Rockfeller Foundation; 2012.
Flick U. Introducción a la Investigación Cualitativa. Ediciones Morata S. L., Madrid; 2004.
Deng X, Yu Z. A Meta-Analysis and Systematic Review of the Effect of Chatbot Technology Use in Sustainable Education. Sustainability. 2023; https://doi.org/10.3390/su15042940
Limna P, Kraiwanit T, Jangjarat K, Klayklung P, Chocksathaporn P. The use of ChatGPT in the digital era: Perspectives on chatbot implementation. J Appl Learn Teach. 2023; https://doi.org/10.37074/jalt.2023.6.1.32
Young DB, Tamir P. Finding Out What Students Know. Sci Teach. 1977;
Tamir P. Self-assessment: the use of self-report knowledge and opportunity to learn inventories. Int J Sci Educ. 1999; https://doi.org/10.1080/095006999290624
Hatim M, Jamil DSI, Khan DSA. Exploring the educational potential of chatGPT. 2023;11.
Okonkwo CW, Ade-Ibijola A. Chatbots applications in education: A systematic review. Comput Educ Artif Intell. 2021; https://doi.org/10.1016/j.caeai.2021.100033
Tlili A, Shehata B, Adarkwah MA, Bozkurt A, Hickey DT, Huang R, et al. What if the devil is my guardian angel: ChatGPT as a case study of using chatbots in education. Smart Learn Environ. 2023; https://doi.org/10.1186/s40561-023-00237-x
Wollny S, Schneider J, Di Mitri D, Weidlich J, Rittberger M, Drachsler H. Are We There Yet? - A Systematic Literature Review on Chatbots in Education. Front Artif Intell. 2021; https://doi.org/10.3389/frai.2021.654924
Tarchi C, Mason L. Effects of critical thinking on multiple-document comprehension. Eur J Psychol Educ. 2020; https://doi.org/10.1007/s10212-019-00426-8
Lu D, Xie Y-N. Critical thinking cultivation in TESOL with ICT tools: a systematic review. Comput Assist Lang Learn. 2024; https://doi.org/10.1080/09588221.2022.2033788
Kasneci E, Sessler K, Küchemann S, Bannert M, Dementieva D, Fischer F, et al. ChatGPT for good? On opportunities and challenges of large language models for education. Learn Individ Differ. 2023; https://doi.org/10.1016/j.lindif.2023.102274
Contreras G. Caracterización del curriculum evaluado en matemática en sexto año básico. Un estudio descriptivo en Valparaíso, Chile. RELIEVE - Rev Electrónica Investig Eval Educ. 2014; https://doi.org/10.7203/relieve.20.2.4295
García-Oliveros G, Salguero-Rivera B, Rodríguez-Díaz O, Palomino-Bejarano E, Caicedo-Valencia R. Las prácticas de evaluación de las matemáticas universitarias: Tensiones y desafíos desde la red conceptual en la que se inscriben. Uniciencia. 2020; https://doi.org/10.15359/ru.34-1.14
Mutiara Sogalrey FA, Madhakomala R, Santosa H, Jamil AI. The Influence of Educators’ Perceptions of Principal Leadership and School Culture on Digital Teacher Literacy Services at School. Al-Tanzim J Manaj Pendidik Islam. 2022; https://doi.org/10.33650/al-tanzim.v6i4.3987

No competing interests reported.

Appendix.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Perceptions of pre-service teachers on the use of ChatGPT during their training and the future implications for their future role in high schools: A research in progress

Status:

Version 1

Abstract

Figures

1 Introduction

2 Theoretical Background

2.1 AI Challenges in Higher Education and Future Teacher Training

2.2 Digital literacy, data literacy, AI literacy, and teacher-mediated AI tools

2.3 Academic Integrity

2.4 Implications for Assessment in Higher Education

3 Methodology

3.1 Research method

3.2 Planning and Implementation Phases

3.2.1 First Phase

3.2.2 Second Phase

3.2.3 Third Phase

3.3 Sample

4 Results

4.1 Prior Knowledge and Student Experience in Using ChatGPT (KPSI)

4.1.1 Knowledge Category

4.1.2 Experience and Use Category

4.1.3 Student Perception Category

4.2 Qualitative Data Analysis

4.2.1 Survey Analysis

4.2.2 Focus Group Analysis

5 Discussion

5.2 Students' perceptions of the implications of ChatGPT in their future role as teachers

6 Conclusion

Declarations

Author Contribution

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1