Exploring Artificial Intelligence Tool Adoption in a Higher Education Faculty’s Pedagogical Practices through CHAT: Supporting International Students in Improving Academic Performance
Download PDF Article
Download Graphical Abstract
Exploring Artificial Intelligence Tool Adoption in a Higher Education Faculty’s Pedagogical Practices through CHAT: Supporting International Students in Improving Academic Performance
Abstract
Artificial Intelligence tools are increasingly utilised across various domains, including higher education, owing to their potential to enhance teaching and learning. This study examines how tutors in a UK higher education faculty adopt AI tools to help international students improve their academic performance. International students often encounter challenges, such as language barriers, cultural differences, and adapting to new educational systems, which can adversely affect their academic success and learning experience. AI tools can potentially mitigate these challenges and support international students. Findings from semi-structured interviews with four tutors reveal the types of AI tools used, factors influencing their adoption, and their perceived effectiveness. The study identifies challenges in integrating AI tools, framed as secondary contradictions through Cultural Historical Activity Theory (CHAT), and offers recommendations for effective integration. These insights can guide faculty policies and training programmes to improve international students’ academic outcomes and enhance AI integration in teaching practices.
Keywords
Artificial Intelligence, higher education, international students, academic performance
JEL Classification
I20, I21
1. Introduction
The application of artificial intelligence (AI) has become increasingly widespread across various sectors, including education, as it is perceived to have the ability to revolutionise processes and enhance efficiency. For example, the implementation of automated grading systems for student assignments can alleviate the workload of tutors, allowing them to allocate more time to instructional activities. Over the past four years, AI has gained recognition as a valuable educational tool, attributed to its perceived benefits in enhancing students’ learning outcomes and providing educators with deeper insights into student progress (Chu et al., 2022). These insights allow tutors to refine their pedagogical practices to help students achieve academic excellence. In the view of Abbas et al. (2023), the possible integration of AI in higher education institutions (HEIs) has gained significant momentum. The adoption and integration of AI in HE presents both benefits and challenges for HEIs and tutors. For international students, AI tools can aid with language and specific feedback that address their distinct requirements and approaches to learning (Popenici & Kerr, 2017). Nevertheless, the authors highlight that the effective adoption and integration of these AI tools requires a thorough examination of their challenges, risks, and ethical considerations.
In response to this growing concern, and recognising the increasing use of AI among students, the HEI in this case study has issued guidance on appropriate AI use in academic practices and encouraged tutors to consider AI tools in their pedagogical approaches. The guidance sets out three main strategies: AI integrated (full adoption and integration of AI tools), AI supported (using AI as a support tool), and AI prohibited (banning the use of AI tools in all aspects of academic practice). As tutors, my colleagues and I have been asked to consider these strategies and determine which ones we will adopt in our pedagogical practices, particularly in relation to assessments. Within my faculty, numerous tutors remain uncertain about the adoption of AI tools due to their lack of experience. Achieving full integration would necessitate a revision of the Intended Learning Outcomes (ILOs) and assessment criteria. Nevertheless, some tutors are adopting AI tools in their pedagogical practices to support international students with the challenges of studying at the UK HEI.
This study examines the adoption of artificial intelligence (AI) tools by the HE faculty tutors to support international students. It focuses on the types of AI tools adopted, the factors influencing their adoption, and their perceived effectiveness in enhancing academic performance. Within the context of this case study, academic performance is defined as the ability of students to meet the entry requirements necessary for progression to the HEI. The international students in this HE faculty are studying pathways such as Foundation, International Year One (equivalent to first-year undergraduate), and Pre-Master’s. Additionally, this study explores the challenges of adopting and integrating AI tools into faculty pedagogical practices. The results of this study have the potential to inform faculty policies and guidelines regarding the integration of AI in teaching. By collecting tutors’ recommendations, the study aims to effectively incorporate AI tools into pedagogical practices and guide the development of training sessions focused on utilising AI tools to enhance the academic performance of international students.
To understand the intricate interaction among tutors, AI tools, and the broader academic context that supports international students, it is essential to employ Cultural Historical Activity Theory (CHAT) in this case study. CHAT offers a comprehensive framework for analysing the tutors’ adoption of AI tools to enhance the academic performance of international students. This case study involves semi-structured interview sessions with four tutors at the start of the academic year. The data collection reflects the tutors’ experience of using AI in their workflow to support international students’ academic performance. The next section outlines the theoretical framework used to underpin this case study, followed by a review of the available literature on this topic, the research design, findings, discussion, and the conclusion of this case study.
This exploration is guided by the following research questions:
- Why and how do tutors adopt AI tools in their teaching practices to support international students’ academic performance?
- What challenges (tensions) do tutors encounter when integrating AI tools to support international students’ academic performance?
2. Theoretical Framework
This case study is grounded in a social constructivist ontology and an interpretivist epistemology. Creswell (2009) states that social constructivists believe that individuals create subjective meanings from their experiences to understand their world, focusing on specific objects or concepts. Driven by these diverse meanings, I explore complex views, rather than limit them to a few categories. This research relies on the participants’ perspectives on the activity system being investigated. From an interpretivist perspective, I consider that people interpret, and act based on their own understanding, and research must account for this (Cohen et al., 2018). This perspective also regards the understanding of the social world as shaped by cultural context and historical positioning (Baxter et al., 2010).
Gedera and Williams (2016) note the growing popularity of CHAT as a multidisciplinary research framework owing to its analytical and conceptual tools for examining human practices. Yamagata-Lynch (2010) highlights that CHAT provides a structure for analysing interactions between people and their environments in real-world contexts. She further explains that CHAT, originating from Russian academics in the 1920s, aims to redefine psychology by examining human activity as an interaction-based engagement with the environment. Consequently, CHAT underpins this case study to explore the interplay between individuals, objectives, tools, group members, and intervening factors in human activities (Koszalka & Wu, 2004). CHAT focuses on the dynamic interactions among these components within an activity system. Clifford (2022) explains that Vygotsky’s initial version of activity theory, based on Marx’s dialectical materialism, posited that human actions are goal-driven and shaped by cultural tools. Individuals use culturally created artefacts to interact with their environment and influence their behaviour and cognition. Activity Theory evolved from the first to the third generation, with Vygotsky’s concept of cultural mediation represented as a triangular relationship among the subject, object, and mediating artefact, (Engeström, 2001) as illustrated in Figure 1.
Figure 1. Vygotsky’s triangle of subject, object and mediating artefact
Source: adapted from Engeström, 2001, p. 134
Engeström (2001) further expounds that the incorporation of cultural objects into human activities has transformed analytical approaches by connecting the Cartesian concept of the individual with broader societal frameworks. This shift necessitates understanding individuals through cultural lenses and examining society through the actions of people who utilise and create cultural artefacts. However, the key constraint of this version is that it focuses on individual-level analysis as the primary unit of study. Mohamad Said et al. (2014) claim that this model was later improved when Engeström expanded upon Leont’ev’s concept of a prehistoric group hunt to present a more comprehensive model of a collective human activity system. The authors further elaborates that this new model incorporates Leontev’s differentiation between individual and group activities. Engeström contended that while Leont’ev made this distinction, he did not explicitly develop Vygotsky’s model into a triangular framework representing a collective activity system as illustrated in Figure 2.
Figure 2. Second generation Activity Theory (AT) triangle
Source: adapted from an original by Engeström, 1997/2014, p. 63
The components in Figure 2 are defined in Table 1 below.
Table 1. Components of CHAT
Source: adapted from Yamagata-Lynch, 2010
In this case study the components of CHAT are represented in Figure 3.
Figure 3. Tutors’ adoption of AI tools to support international students improve their academic performance
In Technology Enhanced Learning (TEL), CHAT is employed, for instance, in the work of Miles (2020) and Núñez (2021). The authors propose that CHAT offers insights into complex scenarios in education and can transform HE. Applying CHAT in this case study offers a robust framework for analysing contradictions and tensions within the activity system. By examining the interactions among subjects, objects, tools, rules, community, and division of labour, CHAT identifies and explores inherent conflicts and inconsistencies. These contradictions may exist as primary tensions within the individual components or as secondary tensions between the components. A systematic investigation of these tensions provides insight into the dynamics of change, innovation, and development within a system. This approach deepens the understanding of how contradictions steer transformation and learning, contributing to the development of practices and the resolution of systemic challenges in adopting AI tools among tutors. Analysing these elements yields deeper insights into the complex interplay and processes involved in AI tool adoption in HEIs.
3. Literature Review
The use of AI tools by students has garnered the interest of researchers, developers, educators, and policymakers (Miao et al. 2021). However, there appears to be a lack of research regarding the use of AI to support international students, which is the central focus of this case study. This case study further investigates the ambiguity surrounding the guidelines or policies related to the use of AI tools among higher education students.
This narrative literature review employs the PRISMA method to search databases such as Google Scholar, ERIC, and Lancaster OneSearch with terms such as ‘teacher AI adoption,’ ‘higher education,’ ‘support international students,’ ‘faculty perspectives of AI,’ and ‘AI in HE policy.’ Literature mentioning ‘schools’ were excluded because of the focus on HEIs. The inclusion criteria were peer-reviewed papers and books in English were included. The initial search yielded thousands of results; therefore, relevance was determined by the titles and abstracts. Only the first ten pages of Google Scholar results were considered, and duplicates from ERIC and Lancaster OneSearch were removed.
Tutors’ AI tool adoption to support international students
Studying in different educational cultures, such as UK HEIs, can be challenging for international students owing to cultural and linguistic differences. To address these challenges, some international students use AI tools to improve their academic performance. Wang et al. (2023) note that HEIs are increasingly exploring AI tools to support international students, offering benefits such as personalised learning, adaptive testing, predictive analytics, and chatbots for learning and research. These tools can enhance learning efficiency and provide tailored educational support. Dotan et al. (2024) acknowledge that AI tools enhance material accessibility and customise educational content for students with diverse linguistic backgrounds, varying levels of preparedness, or different learning preferences, including English language learners.
These perceived benefits motivate tutors to adopt AI to help international students overcome challenges and enhance their academic performance. In the view of Kohnke et al. (2023), tutors can improve pedagogical practices by using generative AI (GenAI) technologies to create and facilitate educational experiences. Tarisayi (2024) posits that AI has the potential to transform HE for international students by personalising teaching, providing formative feedback, identifying at-risk students, and streamlining administrative processes. Luckin et al. (2022) observe that AI tools help tutors make data-informed decisions, reduce workload, and improve classroom management. As stated by Ivanov et al. (2024), tutors invest significant resources in creating educational materials, and AI tools can streamline these tasks to enhance their efficiency. Citing Baidoo-Anu and Ansah (2023), automating educational resource creation reduces workload, allowing more direct student engagement. Al-Mughairi and Bhaskar (2024) mention that ChatGPT functionalities help educators allocate time effectively, enhance lesson delivery, and create a productive educational setting. Another motivating factor is exploration and experimentation with AI tools to improve teaching methods (Al-Mughairi & Bhaskar, 2024). HEIs must equip students with the knowledge to make informed choices regarding AI tools, thus requiring tutors to develop expertise in assessing such usage (Dotan et al., 2024).
Challenges in adopting AI tools in HEIs
Despite the perceived benefits of AI tools in supporting international students improve academic performance, their use of AI tools in HE presents challenges for HEIs, tutors, and students. As outlined in Pedró (2020), while AI has a significant potential to enhance education, its implementation in HEIs has various implications and ethical concerns. This view is echoed by Zhou et al. (2024), who raise concerns regarding international students’ academic integrity and over-reliance on AI.
One of the main challenges in effectively using AI in HEIs to support international students is tutors’ AI readiness. Tutors’ knowledge and confidence in using AI pose another challenge to tutors’ decisions to adopt AI to support international students. Referring to Kohnke et al. (2023), educators should familiarise themselves with these technologies and capabilities before incorporating generative AI tools into their teaching methods. This lack of AI readiness could result in tutors spending more time learning new skills, such as being specific to their prompts to obtain the intended or desired content. Pisica et al. (2023) assert that acquisition of technological proficiency requires extensive training and practice. Therefore, tutors may need support and guidance to effectively incorporate ChatGPT into their teaching methods, as Al-Mughairi and Bhaskar (2024) suggest. In this regard, university faculties must become proficient in applying AI technologies within their respective fields of study (Farelly & Baker, 2023). Even when tutors have AI readiness, they still need to be cautious about the accuracy and reliability of AI-generated content because this content may contain AI hallucinations. In this regard, AI tools can create seemingly plausible information, but in fact, fabrication, as pointed out by Walczak and Cellary (2023). They explain that distinguishing between information and false perceptions can be challenging, because hallucinations are often seamlessly blended with accurate details. Therefore, tutors need to be critical in their decision to use AI-generated content by conducting meticulous fact-checking to ensure that they do not create misleading or inaccurate course content.
Another barrier to AI adoption in classrooms is ensuring equality among students. Access to premium AI tools creates inequality, disadvantaging those who cannot afford them as these versions offer more beneficial functions. Drawing from Wang et al. (2024), a significant number of students lack access to cutting-edge AI technologies, further widening existing educational disparities and technological inequalities. Chan (2023) affirms that ensuring equitable access to AI technologies is imperative for fostering an inclusive learning environment.
Another challenge is unintentional plagiarism as emphasised by Farrokhnia et al. (2023). They affirm that students may utilise ChatGPT because of its promising capabilities without recognising that it may result in plagiarism, and that there is a significant risk of plagiarism becoming more prevalent in academia. This may lead to the assumption that international students are perceived as more prone to academic honesty than home students, as pointed out by Farelly and Baker (2023).
This literature review critically examines the benefits and challenges faced by tutors in their adoption of AI tools to support international students in HE. These include AI readiness, ethical concerns, academic integrity and equitable access. However, there is insufficient empirical research examining the adoption of AI tools to support international students in HEIs. Hence, this study aims to address this shortfall by offering insights and empirical evidence, thereby enhancing our understanding of effective AI tool adoption in pedagogical practices to support international students improve their academic performance.
4. Methodology
Research design
This qualitative case study of four tutors explores the complexities of implementing AI tools in HE to improve international students’ academic performance. Despite the small number of participants, this case study is highly contextualised, and the experiences and perspectives of the tutors are examined closely. Thus, it provides rich insights which larger studies may overlook. Baxter and Jack (2008) note the ability of qualitative case studies to examine complex phenomena within specific contexts, providing rich data and insights into AI tool adoption in HEIs. Yin (2009) argues that case studies are suitable for explaining the ‘how’ and ‘why’ of social phenomena. This method allows for a thorough exploration of the diverse tutor perspectives on AI tool adoption and its impact on pedagogical practices aimed at improving international students’ academic performance. Furthermore, the case study approach aligns with constructivist theory, which asserts that truth varies based on individual perspectives (Baxter & Jack, 2008), reflecting a social constructivist stance. From an interpretivist perspective, investigating a specific case involves interpreting participants’ actions and beliefs and analysing collected information (Bassey, 1999).
However, some criticisms of case studies as a research method include the lack of rigor (Yin, 2009), the inability to generalise the findings, and researcher bias (Tight, 2017). Nevertheless, Adelman et al. (1980 as cited in Bassey, 1999) maintain that case studies allow for generalisations from a specific instance to a larger group, emphasising the distinctions and intricacies of each case. Additionally, case studies recognise the multifaceted and contextual nature of social realities.
Participants
The participants in this case study were four tutors at an HE faculty including two academic managers, a module convener, and a tutor. They all have at least ten years of experience in teaching international students in various subjects, including Academic English. While all participants currently use AI tools, their level of adoption varies based on student needs. This variation in AI tool usage for creating tailored content offers valuable insights into the effectiveness of such tools in improving the academic performance of international students.
Ethical considerations
Ethical procedures were rigorously followed in this case study. The ethics form was approved before the participants were approached. The study adhered to the British Educational Research Association (BERA) guidelines (2024) and provided the participants with an information sheet and consent form. All recordings and transcripts are currently securely stored on the Lancaster University OneDrive. Participants’ names were anonymised during the data analysis stage to ensure privacy and confidentiality, demonstrating a strong commitment to ethical standards.
Data collection
The primary data-collection process involved semi-structured interviews. Anderson and Kanuka (2003) note that semi-structured interviews are guided by specific themes and a theoretical framework. These interviews followed a predetermined guide but allowed flexibility for natural conversation and exploration of relevant tangents (Magaldi & Berler, 2020). In this case study, the questions were formulated based on CHAT elements, with core questions and follow-ups as needed. Each interview lasted approximately one hour and was conducted and recorded in Microsoft Teams for easy transcription. Transcribed data were coded using ATLAS.ti software to enhance the efficiency and depth of analysis.
Data analysis
Interpreting qualitative data can be complex and often lead to multiple analytical perspectives (Cohen et al., 2018). The analysis emphasises detailed and contextual information along with participants’ subjective interpretations. This case study employed thematic analysis to analyse the qualitative data. Braun and Clarke (2017) describe thematic analysis (TA) as a method for identifying, examining, and explaining recurring patterns or ‘themes’ in qualitative data that are adaptable to various theoretical frameworks and research paradigms. The six-phase procedure, as detailed by Kiger and Varpio (2020), includes familiarisation with the data, generating initial codes, identifying themes, reviewing themes, defining and naming themes, and producing the final report. The data was analysed using a hybrid deductive approach, followed by inductive analysis. Bingham (2023) explains that deductive analysis organises data using predefined categories from theoretical frameworks or existing literature, with CHAT used to code the data in this study. Inductive analysis, on the other hand, involves identifying and naming codes and categories through data analysis without prior determination (Bingham, 2023). This approach is valuable in uncovering unexpected themes and providing a contextual understanding of tutors’ experiences with AI tools in teaching.
Findings
The activity system investigated and analysed in this case study is the adoption of AI tools among tutors in an HE faculty. This investigation studied the choice of AI tools being adopted and motivating factors that influenced their adoption pf AI tools, how they are applied in pedagogical practices to support international students, challenges tutors face when using these tools, and possible strategies to integrate AI tools appropriately within the scope of the policy and guidance of the HEI. The four participants identified as P1, P2, P3, and P4 are also addressed as tutors. They have used and are currently using ChatGPT, Claude, Copilot, Grok, and Perplexity in their pedagogical practices to support international students improve academic performance.
Key drivers for adopting AI tools and their application in improving the academic performance of international students
Figure 4 presents the key drivers for tutors’ adoption of AI tools. Although the tutors have different roles such as academic managers, module conveners, and tutors, they all have teaching responsibilities.
Figure 4. Key drivers to adopting AI tools among tutors
The tutors mainly utilised AI tools to support their pedagogical practices, including brainstorming ideas on a particular topic for lecture content, writing and designing course materials, creating model answers and designing questions for assessments, summarising information, and creating lesson plans. The following are examples of tutors’ use of AI tools to support their pedagogical practices.
By utilising AI tools to support their pedagogical practices, the tutors hoped that course materials can help increase student engagement, as mentioned below.
The tutors were motivated to adopt AI tools because of their curiosity about the capabilities of AI tools, personal experience with AI, and the prominence of AI in academic discourse. However, other factors, such as specific job roles and the acceptance of AI in HE, appear to have less influence on AI tool adoption.
Challenges to adopting AI tools
Despite the key drivers mentioned above, the tutors had some reservations and concerns regarding the adoption of AI tools, as presented in Table 2. The concerns articulated by the tutors were informed by their individual experiences with AI tools. No primary contradictions were identified within this activity system; therefore, secondary contradictions are presented in this case study. In the process of identifying secondary contradictions from the interview data, there is a tendency for some overlap among the CHAT elements. The most frequent contradictions that emerged from the interview data are presented herein. The main contradictions within this activity system pertain to the interactions between the Subject and the Tool, and subsequently, how the Tool engages with the Rules, Community, and Division of Labour as presented in Table 2.
| Secondary contradictions | Frequency in interview data | Interview extracts |
| Subject and Tool | 16 | Negative perception of AI tools
P2: Because I didn’t know how to talk to it. And I think I set it up to fail as well. P4: I’m concerned about the internal biases. |
| Tool and Community | 10 | The issue of parity
P3: I think one of the challenges is that the university doesn’t provide an AI package that is accepted and standardises student access to AI. So, you’ve got kind of a situation where students who are more tech savvy or who have the finances to buy the premium versions of AI packages potentially have an advantage over the students that don’t know about them or cannot afford to use them. So, you’re creating another area of inequality within the student body as well. Prejudice towards international students P1: So, you know, home students. Yes. You can go off and use AI, don’t worry about it. And international students you can’t because, we’re concerned about you using your own language. … it’s something that international students are going to be very, very sensitive about. P3: … it could perpetuate kind of a distrust of international students, work a little bit more, even though they’re probably doing the same things … |
| Tool and Division of Labour | 9 | Increased workload
P1: The end product, even if you’ve refined it several times and you’ve gone back and … even when you’ve done that, you’ve then got to look at what has been created and adapt and edit to make it suitable for use. The need to upskill P2: … it took me a long time to work out what language I need to use in order to get the AI to produce the material for me. P4: … if you get the wrong prompt or too general a prompt, you’re basically driving the wrong way down a one-way street and that never ends well. |
| Tool and Object/Outcome | 8 | Over reliance on AI tools
P1: One of the main focuses of these programmes is to develop the skills in the language skills of these students, … AI tools can provide a shortcut for students to produce a piece of work that doesn’t necessarily reflect their own skills or language competencies, since the piece of work might be generated by the AI tool itself. P2: In the later stages, I think that it can be a hindrance. I think if they’re looking to it for phrases, chunks of language, then that’s when they get into problems. |
| Rules and Tool | 8 | Lack of clear policies/guidance
P1: Everybody’s aware of AI tools now that they and almost all students are using them to some degree or other, and that I think, is why it’s so important to have a kind of clear policy … to ensure that these tools are being used in a proper way that doesn’t undermine the credibility of English language programmes. Ethics concerning intellectual property P3: … they’re putting intellectual property of the university into a generative AI. Should they do that or not is a question of ethics … |
Table 2. Frequency of secondary contradictions in CHAT and supporting evidence from interview data
The contradiction between the Subject and Tool arises when the tutors expressed negative perceptions of AI tools and concerns about inherent biases, reflecting a disconnect between AI’s intended purpose to assist teaching and tutors’ scepticism about its reliability and fairness. Concerning the tension between the Tool and Community, tutors highlight issues such as non-standardised access to AI tools, varying levels of experience and knowledge, and the advantages of premium versions. Additionally, there is potential prejudice against international students suspected of using AI in coursework. The contradiction between the Tool and Division of Labour highlights the misalignment between the skills needed to use AI tools effectively and tutors’ knowledge. The interview data also revealed a contradiction between the Tool and Object/Outcome, where tutors noted that international students’ over-reliance on AI tools could hinder the development of essential academic skills. The lack of clear AI policies or guidance created tension between the Rules and Tool, revealing uncertainty about integrating AI tools while maintaining academic integrity and protecting intellectual property rights.
The identified secondary contradictions reveal conflicts and inconsistencies within the activity system, highlighting potential areas for enhancement to ensure the ethical and effective utilisation of AI tools in assisting international students in improving their academic performance.
5. Discussion
The findings provide several important insights into the adoption and utilisation of AI tools by tutors in the HE faculty. AI tools are primarily utilised to provide pedagogical support to tutors and are increasingly perceived as a valuable resource to support pedagogical practices, as indicated by all four tutors. Pedagogical support involves decreasing tutors’ workload (Luckin et al., 2022; Baidoo-Anu & Ansah, 2023). The AI tools’ capability of providing personalised learning to increase student engagement, which was cited by three tutors, aligns with Al-Mughairi and Bhaskar’s (2024) theme of personalised teaching and learning, and Tanveer et al.’s (2020) observation that AI tools promote personalised learning and improve learning outcomes. This allows tutors to focus on delivering high-quality teaching and mentoring to international students, thereby leading to a more supportive and engaging learning environment.
The data also exposed an interesting pattern in which tutors are influenced by intrinsic factors, such as their curiosity about the capabilities of AI tools and personal experience. These intrinsic motivating factors seem to be more influential than institutional and attitudinal factors such as job roles and acceptance of AI tools. Radhakrishnan and Chattopadhyay (2020) propose that intrinsically motivated individuals feel that they can control their learning and utilisation of AI tools, resulting in a more creative and successful implementation of the technology.
In contrast, institutional and attitudinal factors tend to be less engaging on a personal level, as they often lack the direct and personal connections offered by intrinsic factors. These findings suggest that effective AI integration in HE to support international students might require a complex strategy that addresses both practical teaching needs and personal development opportunities, while simultaneously promoting an environment of open academic discourse about AI technologies.
Nevertheless, the tutors faced challenges in their adoption of AI tools, which are translated into secondary contradictions in Table 2. These contradictions seem to emerge from the limitations of AI tools and the lack of rules guiding the use of these tools, thereby impacting the division of labour in the activity system, as seen in Figure 5.
Figure 5. Secondary contradictions in the activity system
Subject AND Tool
The tutors had negative perceptions of the AI tools because they were not familiar with the functions and capabilities of the AI tools. Additionally, because the negative use of AI tools among students is frequently highlighted in academic discourse, tutors already had a negative perception of AI tools even before they tried using the tools themselves. In this regard, it is vital to address the concerns raised by academic tutors and improve AI transparency to foster the acceptance of AI tools, leading to their effective integration in HE. Internal biases within AI tools have also caused contention, as shown in the interview data in Table 2. This tension is also echoed by Farrokhnia et al. (2023), as ChatGPT could potentially enable discrimination and biases in education. Hence, tutors and students should receive training and guidance to critically assess AI-generated content (Wang et al. 2024). Nonetheless, organising training sessions for tutors and students may not always be practical, as such initiatives demand time, and the development of an AI literacy module requires careful consideration.
Tool AND Community
This contradiction manifests mainly in the issue of parity when international students use AI tools in their academic practice. This could lead to a wider digital divide among international students. This concern emphasises AI integration in HE to be carefully considered to ensure that all students are treated equally, regardless of their background. Therefore, training in AI literacy should be included in the HE curriculum, instead of discouraging the use of AI tools. In support of this, Wang et al. (2024) encouraged the incorporation of these tools because it is more likely to yield favourable outcomes. Another issue is the negative perception of international students by the wider academic community, owing to their language proficiency and accuracy. To compensate for their seemingly low English proficiency, international students would resort to using AI tools in their academic practices so that they can be on par with home students. This situation could lead to discriminatory practices towards international students who feel that they are unfairly judged or targeted and scrutinised, resulting in a lack of respect between international students and the wider academic community. According to Farrelly and Baker (2023), international students are likely to be more frequently and unjustly suspected of misusing AI tools than their local peers are, highlighting significant fairness and equality concerns. This erosion of trust could also negatively impact international students’ mental health and well-being.
Tool AND Division of Labour
Despite AI tools being used to optimise workflow, the tutors had to constantly adapt and refine course materials to ensure that these materials can support efficient engagement with course materials by international students in order to enhance their academic performance. Tutors also mentioned the need to upskill and learn new ways of communicating with AI tools to assist them in their workflow. For instance, the interview extracts highlight that the constant need to refine course materials seems to add more stages to the workflow of tutors, and the extra effort and time to communicate with AI tools through prompts also increases workload. Continuous refinement of the prompts is necessary to achieve the desired outcomes. These challenges do not appear to reflect the claim that AI tools, such as ChatGPT, could decrease teachers’ workloads, as reported by Farrokhnia et al. (2023). This contradiction accentuates the need to address technological implementation and tutors’ skill development to ensure appropriate integration of AI tools within current academic practices. Dotan et al. (2024) assert that tutors’ AI readiness is important to ensure students are equipped with the knowledge and skills to use AI tools appropriately. However, achieving a proficient level of AI readiness requires training and effort (Pisica et al., 2023). To address the issue of AI readiness and training among tutors, the HEI can organise training sessions focused on the development of specific prompts. Additionally, tutors with experience in utilising AI tools could provide mentorship to those less familiar with these technologies.
Tool AND Object/Outcome
Another contradiction that emerged in the interview data is that the utilisation of AI tools could contribute to international students relying too much on them that the necessary academic and study skills are not developed; hence, the object or outcome may not be attained. Over-reliance on AI could potentially undermine international students’ independent learning and skill acquisition, thus making them ill-equipped with the necessary skills that are crucial for academic success, and possibly in their professional life after graduation. The use of AI tools in this manner could result in a superficial understanding of the disciplines being studied, rather than deep, meaningful learning, and profoundly impact academic integrity. According to Wang et al. (2024), students must dedicate time and effort to gaining the necessary skills and knowledge by employing legitimate methods to maintain academic integrity. One strategy to mitigate this over-reliance is to have open discourse and share effective methods to harness the capabilities of AI tools within acceptable parameters.
Rules AND Tool
The lack of clear policies and guidance has resulted in tutors and international students not knowing how to use AI tools within the parameters permitted by HEIs. This can lead to confusion and inconsistencies in the use of AI in HEIs, resulting in different standards. Additionally, this issue could lead to the potential misuse or unintended misuse of AI in coursework. This could be attributed to the implementation of AI policies in HEIs, which are still in their early stages (Tanveer et al., 2020) and can be mitigated by HEIs collaborating to produce a set of standard policies that can be adapted by individual HEIs to suit their educational contexts. Additionally, the assessment focus should shift to include more portfolio work and seminars to discuss written work produced by international students. Tutors need to have a more positive outlook on utilising AI by changing their mindset and adopting a more positive attitude towards AI tools so that they can support international students in improving their academic performance. Chan (2023) suggests that comprehensive training programmes must be implemented for both students and educators to effectively utilise and incorporate AI technologies into teaching and learning practices. Additionally, clear policies and guidelines on the ethical use of AI should be established to address the opportunities and challenges presented by AI tools.
In conducting this case study, it would have been useful to interview students about their use of AI tools and observe their application in classroom settings to gauge their effectiveness in improving international students’ academic performance. The lack of student perspectives in this investigation limits the understanding of the direct impact of AI tools on international students’ academic performance. Utilising CHAT in this case study limits the identification of contradictions to secondary ones, which often overlap between the elements, complicating their specific identification. Despite numerous secondary contradictions in this activity system, only the most frequent contradictions are discussed in this paper. Another limitation is the absence of primary contradictions. Future research could address this issue by combining semi-structured and unstructured interviews to expose the primary contradictions.
6. Conclusion
This case study elucidates the primary factors influencing tutors’ adoption of AI tools. These tools appear to have facilitated tutors’ pedagogical practices, encompassing activities such as information retrieval akin to a search engine, the generation of course materials, the creation of content and questions for assessments, and the optimisation of workflow. Nonetheless, the integration of AI tools into educational practices presents challenges, including inefficient workflows due to the limitations of AI tools and issues of academic integrity stemming from the absence of explicit policies or guidelines regarding the appropriate use of AI among international students. This situation is perceived to have engendered bias against international students. These challenges are identified as secondary contradictions when analysed through CHAT, with emphasis on the most frequent contradictions. These are categorised as Subject and Tool; Tool and Community; Tool and Division of Labor; Tool and Object/Outcome; and Rules and Tool. The study discusses these contradictions and proposes strategies to mitigate them, thereby fostering the positive transformation of the activity system. The application of CHAT in this research offers valuable insights into the factors affecting tutors’ adoption of AI tools, their efficacy in supporting international students, and the challenges encountered by tutors in integrating these tools into pedagogical practices. This research highlights the potential of AI tools to enhance international students’ academic performance by addressing specific challenges, such as language barriers and cultural differences. Furthermore, the study provides actionable recommendations for effective AI integration and informs faculty policies and training programmes aimed at improving educational practices and outcomes. These findings contribute to our understanding of AI tool adoption and its role in promoting inclusive teaching practices in higher education. To address the absence of primary contradictions in this study, future research could investigate these within the context of AI adoption in educational environments. Conducting longitudinal studies may reveal emerging contradictions as institutions integrate these AI tools. Alternatively, ethnographic research could be employed to observe the utilisation of AI tools in classroom settings.
Acknowledgement
This research was undertaken as part of the PhD in e-Research and Technology Enhanced Learning in the Department of Educational Research at Lancaster University. I am pleased to acknowledge the contribution of tutors and peers in supporting the development of this study and its report as an assignment paper.
About the Author
Felicia Heard
Lancaster University, United Kingdom
References
Abbas, N., Imran, A., Manzoor, R., Hussain, T., & Al Hussaini, M.H. (2023). Role of Artificial Intelligence Tools in Enhancing Students’ Educational Performance at Higher Level. Journal of Artificial Intelligence, Machine Learning and Neural Network, 3(5), pp. 36-49. https://doi.org/10.55529/jaimlnn.35.36.49.
Al-Mughairi, H. & Bhaskar, P. (2024). Exploring the factors affecting the adoption AI techniques in higher education: insights from teachers’ perspectives on ChatGPT. Journal of Research in Innovative Teaching & Learning, pp. 1-16. https://doi.org/10.1108/JRIT-09-2023-0129.
Anderson, T. & Kanuka, H. (2003). e-Research: Methods, Strategies, and Issues. Pearson Education, Inc.
Baidoo-Anu, D. & Owusu Ansah, L. (2023). Education in the Era of Generative Artificial Intelligence (AI): Understanding the Potential Benefits of ChatGPT in Promoting Teaching and Learning. Journal of AI, 7(1), pp. 52-62. https://doi.org/10.61969/jai.1337500.
Bassey, M. (1999). Case Study Research in Educational Settings. Open University Press.
Baxter, P. & Jack, S. (2008). Qualitative Case Study Methodology: Study Design and
Implementation for Novice Researchers. The Qualitative Report, 13(4), pp. 544-559. http://www.nova.edu/ssss/QR/QR13-4/baxter.pdf.
Baxter, L., Hughes, C. & Tight, M. (2010). How to research (4th ed.). Open University Press.
Bingham, A.J. (2023). From Data Management to Actionable Findings: A Five-Phase Process of Qualitative Data Analysis. International Journal of Qualitative Methods, 22, pp. 1-11. https://doi.org/10.1177/16094069231183620.
Braun, V. & Clarke, V. (2017). Thematic analysis. The Journal of Positive Psychology, 12(3), pp. 297-298. https://doi.org/10.1080/17439760.2016.1262613.
British Educational Research Association (2024). Ethical Guidelines for Educational Research (5th ed.). https://www.bera.ac.uk/publication/ethical-guidelines-for-educational-research-fifth-edition-2024.
Chan, C.K.Y. (2023). A comprehensive AI policy education framework for university teaching and learning. International Journal of Educational Technology in Higher Education, 20(38), pp. 1-25. https://doi.org/10.1186/s41239-023-00408-3.
Chu, H.C., Hwang, G.H., Tu, Y.F., & Yang, K.H. (2022). Roles and research trends of artificial intelligence in higher education: A systematic review of the top 50 most-cited articles. Australasian Journal of Educational Technology, 38(3), pp. 22–42. https://ajet.org.au/index.php/AJET/article/view/7526#:~:text=Higher%20education%20practitioners%20and%20researchers%20need%20to%20be%20aware%20of.
Clifford, S. (2022). The use and value of cultural historical activity theory in institutional educational technology policy. Studies in Technology Enhanced Learning, 3(1), pp. 1-29. https://doi.org/10.21428/8c225f6e.06989392.
Cohen, L., Manion, L. & Morrison, K. (2018). Research Methods in Education (8th ed.). Routledge.
Creswell, J.W. (2009). Research Design: Qualitative, Quantitative, Mixed Methods Approaches. Sage Publications, Inc.
Dotan, R., Parker, L.S. & Radzilowicz, J.G. (2024, June 3-6). Responsible Adoption of Generative AI in Higher Education: Developing a “Points to Consider” Approach Based on Faculty Perspectives. The 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’24), Rio de Janeiro, Brazil. https://doi.org/10.1145/3630106.3659023.
Engeström, Y. (2001). Expansive Learning at Work: toward an activity theoretical reconceptualization. Journal of Education and Work, 14(1), pp. 133-156. https://www.tandfonline.com/doi/abs/10.1080/13639080020028747.
Engeström, Y. (1997, 2014). The Emergence of Learning Activity as a Historical Form of Human Learning (25-108). In Learning by Expanding: An Activity-Theoretical Approach to Developmental Research (2nd ed.). Cambridge University Press. https://doi.org/10.1017/CBO9781139814744.
Farelly, T. & Baker, N. (2023). Generative Artificial Intelligence: Implications and Considerations for Higher Education Practice. Education Science, 13 (1109), pp. 1-14. https://doi.org/10.3390/educsci13111109.
Farrokhnia, M., Bannihashem, S.K., Noroozi, O. & Wals, A. (2023). A SWOT analysis of ChatGPT: Implications for educational practice and research. Innovations in Education and Teaching International, 61(3), pp. 460-474. https://doi.org/10.1080/14703297.2023.2195846.
Gedera, D.S.P. & Williams, P.J. (eds) (2016). Activity Theory in Education: Research and Practice. Sense Publisher. https://link.springer.com/book/10.1007/978-94-6300-387-2.
Ivanov, S., Soliman, M. Tuomi, A., Alkathiri, N.A., & Al-Alawi, A.N. (2024). Drivers of generative AI adoption in higher education through the lens of the Theory of Planned Behaviour. Technology in society, 77(10251), 1-14. https://doi.org/10.1016/j.techsoc.2024.102521.
Kiger, M. E., & Varpio, L. (2020). Thematic analysis of qualitative data: AMEE Guide No. 131. Medical Teacher, 42(8), pp. 846–854. https://doi.org/10.1080/0142159X.2020.1755030.
Kohnke, L., Moorhouse, B.L. & Zou, D. (2023). Exploring generative artificial intelligence preparedness among university language instructors: A case study. Computers and Education: Artificial Intelligence, 5(100156), pp. 1-8. https://doi.org/10.1016/j.caeai.2023.100156.
Koszalka, T.A. & Wu, C. (2004). A cultural historical activity theory [CHAT] analysis of technology integration: Case study of two teachers. https://www.researchgate.net/publication/251434047_A_cultural_historical_activity_theory_CHAT_analysis_of_technology_integration_Case_study_of_two_teachers.
Luckin, R., Cukurova, M., Kent, C. & du Boulay, B. (2022). Empowering educators to be AI-ready. Computers and Education: Artificial Intelligence, 3(100076), pp. 1-11. https://doi.org/10.1016/j.caeai.2022.100076.
Magaldi, D. & Berler, M. (2020). Semi-structured Interviews. In Zeigler-Hill, V. & Shackelford, T.K. (Eds). Encyclopedia of Personality and Individual Differences, pp. 4825–4830. Springer. https://doi.org/10.1007/978-3-319-24612-3_857.
Miao, F., Holmes, W., Huang, R., Zhang, H. & UNESCO (2021). AI and education: A guidance for policymakers. UNESCO Publishing. https://doi.org/10.54675/PCSP7350.
Miles, R. (2020). Making a case for Cultural Historical Activity Theory: Examples of CHAT in practice. Studies in Technology Enhanced Learning, 1(1). https://stel.pubpub.org/pub/01-01-miles-2020/release/1.
Mohamad Said, M.N.H., Foret, M. & Eames, C. (2014). Analysis of Contradictions in Online Collaborative Learning using Activity Theory as Analytical Framework. Jurnal Teknologi, 68(2), pp. 57-63. http://dx.doi.org/10.11113/jt.v68.2910.
Núñez, A.-M. (2021). Applying Cultural Historical Activity Theory (CHAT) perspectives toward equity in higher education organizations and systems. In L. W. Perna (Ed.), Higher education: Handbook of theory and research (37). Springer. https://doi.org/10.1007/978-3-030-66959-1_10-1.
Pedró, F. (2020). Applications of Artificial Intelligence to higher education: possibilities, evidence, and challenges. IUL Research, 1(1), pp. 1-16. http://dx.doi.org/10.57568/iulres.v1i1.43.
Pisica, A.I., Edu, T., Zaharia, R.M. & Zaharia, R. (2023). Implementing Artificial Intelligence in Higher Education: Pros and Cons from the Perspectives of Academics. Societies, 13(5), pp. 1-13. https://doi.org/10.3390/soc13050118.
Popenici, Ş. & Kerr, S. (2017). Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning, 12(22), pp. 1-13. https://doi.org/10.1186/s41039-017-0062-8.
Radhakrisnan, J. & Chattopadhyay, M. (2020). Determinants and Barriers of Artificial Intelligence Adoption – A Literature Review. In Sharma, S.K., Dwivedi, Y.K., Metri, B. & Rana, N.P. (Eds). Re-imagining Diffusion and Adoption of Information Technology and Systems: A Continuing Conversation. TDIT 2020. IFIP Advances in Information and Communication Technology, 617. Springer. https://doi.org/10.1007/978-3-030-64849-7_9.
Tanveer, M., Hassan, S. & Bhaumik, A. (2020). Academic Policy Regarding Sustainability and Artificial Intelligence (AI). Sustainability, 12(22), 9435. https://doi.org/10.3390/su12229435.
Tarisayi, K. S. (2024). Strategic leadership for responsible artificial intelligence adoption in higher education. CTE Workshop Proceedings, 11, pp. 4–14. https://doi.org/10.55056/cte.616.
Tight, M. (2017). Understanding Case Study Research: Small-scale Research with Meaning. SAGE Publications Ltd. https://doi.org/10.4135/9781473920118.
Walzcak, K. & Cellary, W. (2023). Challenges for higher education in the era of
widespread access to Generative AI. Economics and Business Review, 9(2), pp. 71-100. https://doi.org/10.18559/ebr.2023.2.743.
Wang, T., Marengo, A., Lund, B.D., Pange, J., Teel, Z.A., Pagano, A. & Mannuru, N.R. (2023). Exploring the Potential Impact of Artificial Intelligence (AI) on International Students in Higher Education: Generative AI, Chatbots, Analytics, and International Student Success. Applied Sciences, 13(11), 6716. https://doi.org/10.3390/app13116716.
Wang, N., Wang, X. & Su, Y. (2024). Critical analysis of the technological affordances, challenges and future directions of Generative AI in education: a systematic review. Asia Pacific Journal of Education, 44(1), pp. 139-155. https://doi.org/10.1080/02188791.2024.2305156.
Yamagata-Lynch, L.C. (2010). Activity Systems Analysis Methods: Understanding Complex Learning Environments. Springer. https://link.springer.com/chapter/10.1007/978-1-4419-6321-5_2.
Yin, R.K. (2009). Case Study Research: Design and Methods (4th ed.). Sage Publications Inc.
Zhou, X., Zhang, J., & Chan, C. (2024). Unveiling Students’ Experiences and Perceptions of Artificial Intelligence Usage in Higher Education. Journal of University Teaching and Learning Practice, 21(6), pp. 1-20. https://doi.org/10.53761/xzjprb23.
Gallery
