Exploring intention of undergraduate students to embrace chatbots: from the vantage point of Lesotho

Researchers and stakeholders have acknowledged the significance of integrating Industry 4.0 technologies into higher education, leading to initiatives aimed at their promotion within higher education institutions (HEIs) (Ayanwale et al., 2022; Ayanwale, 2024). Early exposure to these emerging technologies is deemed crucial for fostering students' competencies and establishing a solid foundation for future learning and career development (Touretzky & Gardner-McCune, 2022; Kim et al., 2023). Coined to describe the fourth industrial revolution (4IR), Industry 4.0 involves the integration of digital technologies into global industrial processes and economies (Ayanwale, 2023a; Mamadou & Ernesto, 2020). This encompasses the Internet of Things (IoT), artificial intelligence (AI), cloud computing, virtual and augmented reality, big data, and robotics. Within higher education (HE), the adoption of Industry 4.0 technologies is propelled by various factors, including preparing students for the contemporary workplace, enhancing the quality of teaching and learning, and ensuring competitiveness in a globalized environment (Almela, 2023). This study specifically focuses on a pivotal technology within Industry 4.0—artificial intelligence (AI), increasingly employed in HEIs to support students in their academic pursuits. AI refers to machines demonstrating human-like intelligence, capable of executing tasks and undergoing self-improvement through the collection and storage of data (Yau et al., 2022). AI application tools, including ChatGPT, Bing Chat, paperpal, paperdigest, Elicit, Humata, Copilot, trinkin.ai, Paper Brain, and ExperAI, are actively used to streamline student assignments, research, and term papers, facilitating and expediting these processes. Comprehending how AI operates and the problems it can address is crucial for students to effectively curate the tools they use in their lives (Ayanwale, 2023; Ma et al., 2023). Ongoing efforts aim to develop curriculum frameworks, guidelines, and standards to effectively introduce AI application tools and concepts to students in HE. The integration of these technologies into the HE curriculum is anticipated to enhance learning outcomes, elevate student engagement and motivation, and foster the development of digital skills increasingly sought after in the job market (Mamadou & Ernesto, 2020; Oladele et al., 2022, 2023).

The exploration of emerging technologies, particularly the adoption of AI application tools such as chatbots, is a crucial subject of discourse among higher education (HE) students. Understanding the psychological factors that shape students' decisions to embrace these technologies is vital for devising effective strategies to support their adoption. However, existing studies predominantly focused on students' adoption of chatbots in HE outside the African context, employing frameworks like the Technology Acceptance Model (TAM), Unified Theory of Acceptance and Use of Technology (UTAUT), or Theory of Planned Behaviour (TPB) (Roy et al., 2022; Pillai et al., 2023; Garcia de Blanes et al., 2022; Alboqami, 2023; Malik et al., 2021; Mohd Rahim, et al., 2022). The generalizability of these findings to other regions is questionable. Moreover, a prior study by Ragheb et al. (2022) exclusively focused on students from private universities in Egypt, limiting its applicability to other regions. In light of these gaps, the present study aims to extend Ragheb et al.'s (2022) research by examining the behavioral intentions of public undergraduate students in a Southern African country regarding the adoption of AI application tools (chatbots) for academic purposes. To achieve this, we leverage the Innovation Diffusion Theory (IDT) proposed by Rogers (1995, 2003), a widely recognized theory utilized to comprehend the acceptance and adoption of new technologies. IDT has found application in various studies exploring the adoption of AI application tools (Aizstrauta et al., 2014; Al-Rahmi et al., 2019; Pinho et al., 2021). Furthermore, perceived trust in using chatbots emerges as a significant factor influencing students' behavioral intention to engage with them in the context of higher education (Al-Sharafi et al., 2016; Laumer et al., 2019; Mohd Rahim, et al., 2022; Ragheb et al., 2022).

In summary, existing research on chatbots adoption in higher education in Northern Africa, particularly at a private university in Egypt (Ragheb et al., 2022), is confined to specific contexts. Given Africa’s diverse nature and distinct needs for embracing AI tools, it becomes imperative to explore the viewpoints of undergraduate students in other regions, such as Southern Africa. Considering the unique social context of each region is essential for devising strategies to introduce and support the adoption of AI application tools among students in higher education. Therefore, this study aims to contribute to the expanding literature on how undergraduate students perceive the adoption of AI tools in higher education. It particularly focuses on public higher education undergraduate students across various disciplines, including Science, Technology, Education, Humanities, Health Sciences, Social Sciences, Agriculture, and Law. Examining their behavioral intentions to use AI tools provides valuable insights for supporting adoption in higher education and enhancing student learning outcomes.

In this research paper, we present the findings of a quantitative study involving 842 undergraduate students from different faculties at the National University of Lesotho, investigating their behavioral intentions to use chatbots for academic purposes. The study explores various psychological factors such as relative advantages, observability, trialability, perceived usefulness, perceived ease of use, perceived trust, and behavioral intention to use AI applications. The primary research question addressed is, "How do the seven psychological factors relate to the intention to use AI applications by higher education students?" The paper's structure includes an introduction and rationale, a literature review providing context and emphasizing the importance of AI tool adoption among higher education students in Lesotho, a section detailing the research methodology, and sections presenting and interpreting the study's results. The paper concludes by identifying study limitations and proposing directions for future research.

The incorporation of emerging technologies, such as AI, has significantly improved the landscape of teaching, learning, and research at universities. While AI adoption in higher education (HE) is on the rise in Lesotho and Africa as a whole, it lags behind more developed regions, particularly in the global north. Numerous studies have explored the adoption and usage of AI application tools, particularly chatbots, in various contexts. These contexts range from customer service chatbots (Sheehan, 2018) to Siri chatbot implementation in tourism (Ivanov et al., 2019), and enterprise chatbots (Brachten et al., 2021). In the realm of education, research has delved into AI-based T-bot adoption in higher education (Pillai et al., 2023), chatbots in higher education (Ragheb, et al., 2022), AI-based applications in higher education (Wang & Tu, 2021), social chatbots supporting students for studies (Al-Ghadhban & Al-Twairesh, 2020), the role of chatbots in education (Molnar & Szuts, 2018), chatbots for learning (Malik et al., 2021), chatbots for e-learning (Colace et al., 2018), and Nabhiha—the role of chatbots in education (Molnar & Szuts, 2018). These studies collectively demonstrate the potential of chatbots to enhance teaching, learning, and research in the field of education.

While existing research provides valuable insights into the factors influencing chatbot adoption and use in education, there is a gap in understanding the specific factors affecting chatbot adoption and use in HE, particularly in developing countries like Lesotho. This study seeks to address this gap by exploring the factors that influence university students' behavioral intention to adopt AI application tools, specifically chatbots, in higher education. Identifying these factors is crucial for Lesotho and other developing countries to devise strategic initiatives that effectively promote the integration of AI in teaching, learning, and research. By understanding the influences on students' behavioral intentions to adopt chatbots for academic purposes, developing countries can establish a more comprehensive framework for comprehending the adoption and diffusion of technological innovations in higher education.

This study focused on the expanded diffusion innovation theory and its application to the adoption of AI chatbot tools for academic purposes among undergraduate students from a public university in Lesotho. We examined factors such as perceived ease of use, perceived usefulness, and perceived trust in relation to students' behavioral intention to use chatbots. We also explored individual factors like relative advantage, perceived compatibility, and perceived trialability and how they influenced students' willingness to adopt AI technology. The study found significant results indicating the relationships between these factors and students' intention to use chatbots. These findings contribute to our understanding of the factors that influence the adoption of AI tools, specifically chatbots, in educational settings in the Southern part of Africa. The study revealed that there is a positive and significant relationship between the perceived relative advantages of chatbots and students' behavioral intention to use them. This suggests that students recognize the benefits of using AI tools and are willing to incorporate them into their academic activities. This finding aligns with the innovation diffusion theory, which suggests that individuals are more likely to adopt an innovation if they perceive it as advantageous (Nguyen & Nguyen, 2020; Rogers, 2003). The relative advantages of chatbots, such as personalized assistance and timely responses, contribute to their perceived value for students in their academic pursuits. Furthermore, the study found that compatibility between individuals and chatbots has a direct and significant effect on behavioral intention. Students who perceive a sense of compatibility with AI tools are more likely to embrace them as useful tools for learning. This finding is consistent with the expanded innovation diffusion theory, which emphasizes the importance of compatibility in the adoption process of new technology (Rogers, 2003). When students perceive chatbots as compatible with their existing learning practices and preferences, they are more inclined to use them. This study also confirms the findings by Shin et al. (2020) and Olasina (2019) that individuals perceived greater compatibility with AI tools, which led them to interact and engage with the chatbot more frequently. As a result, users are more likely to engage in desired behaviors, such as continued usage or adoption of the chatbot, if they have a positive user experience and a sense of alignment with its capabilities, personality, and communication style.

The study also examined the relationship between the perceived trialability of chatbots and behavioral intention. The findings supported this hypothesis, indicating that students who perceive the opportunity to try out AI tools are more likely to have the intention to use chatbots for learning purposes. This aligns with previous research that highlights the significance of trialability as a means to reduce uncertainty and encourage adoption (Almeida, 2023, Pinho et al., 2021; Rogers, 1983, 2003). The perceived trialability of chatbots allows students to experiment with the tool and experience its potential benefits, thus increasing their intention to use it. The presence of other factors, however, could raise doubts about this relationship. Several studies have shown that perceived usefulness and ease of use are key factors determining users' behavior toward technology (Davis et al., 1989; Laumer et al., 2019). Potentially, these factors might overwhelm perceived trialability's influence on behavioral intentions. Other factors, such as perceived usefulness or ease of use, may reduce or mediate a preference for trialability's significance. Moreover, the level of trust that students have in AI tools was found to play a role in shaping their intention to use chatbots for academic purposes. Trust is a critical factor in the adoption of new technologies, as individuals are more likely to adopt innovations they perceive as trustworthy (Aoki, 2020; Venkatesh et al., 2012). In the context of AI application tools, students’ trust in the tool’s ability to provide accurate information and support their learning needs influences their intention to use it (Tarhini et al., 2017).

Additionally, a positive and significant correlation exists between perceived usefulness and perceived ease of use, suggesting students find AI tools useful and easy to use. In the adoption process of emerging technologies, perceived usefulness and ease of use play an important role, as noted by Malik et al. (2021) and Cao et al. (2021). Student adoption of and integration of chatbots into their learning activities is enhanced when they perceive chatbots as useful and easy to use. The study also examined the relationships between perceived usefulness, perceived ease of use, and students' behavioral intention to use chatbots. However, the findings did not support the hypotheses proposing direct effects of perceived usefulness and perceived ease of use on behavioral intention. These results are somewhat surprising, as previous research has consistently shown that perceived usefulness and ease of use are strong predictors of intention to use technology (Albanna et al., 2022; Davis et al., 1989; Huang et al., 2021; Tselios et al., 2011). It is important to consider the study's specific context and the sample population's characteristics when interpreting these results. Firstly, the study's unique context, which involved undergraduate students in Lesotho, may introduce cultural, educational, or technological differences that can influence how students perceive the relationships between PU, PEU, and BI. These specific contextual factors might shape perceptions differently compared to the contexts examined in previous research. The nature of academic tasks could also play a role in shaping students' perceptions. Depending on task complexity or personalization requirements, students might prioritize factors like trialability or compatibility over PU and PEU when it comes to influencing their behavioral intentions. Additionally, the study population's limited familiarity with chatbots may impact how they perceive such technology's usefulness and ease of use. Students might rely more on factors like trust or compatibility due to a lack of firsthand experience with chatbots. The interaction effect between PU and PEU could contribute to the unexpected results. Instead of their individual effects, the combined influence of these two factors might shape students' behavioral intentions. The study did not explicitly explore interaction effects, so there is room for further investigation. Lastly, sample characteristics, including technological literacy, prior exposure to AI technologies, and individual preferences, may influence the observed relationships. If the sample exhibits specific characteristics that differ from other populations in the literature, this could contribute to the lack of support for the hypotheses. Further, the results of the multiple group analysis reveal gender-specific variations in the factors influencing the behavioral intention to adopt chatbots among undergraduate students in Lesotho. Male students show a stronger positive association between relative advantage and behavioral intention, emphasizing the importance of perceived benefits for males. Additionally, compatibility has a more influential role for males in their adoption intention. Both genders consider trialability important, with slight differences. Trust is crucial for both, slightly more so for males, while perceived usefulness plays a more decisive role for females. Female students also show a stronger relationship between perceived usefulness and ease of use, suggesting its significance in shaping adoption intention for females. Overall, recognizing these gender-specific differences is vital for tailoring effective strategies and interventions for encouraging chatbot adoption among students.

The practical implications of this study's findings are relevant for undergraduate students in higher education, specifically in Lesotho. By understanding the implications, students can make informed decisions about the use of chatbot technology and effectively integrate it into their learning processes. Incorporating chatbot technology into higher education has the potential to enhance students’ learning experiences and support their academic success. However, students should approach the adoption of chatbot technology thoughtfully, considering the benefits, compatibility, trial opportunities, trustworthiness, usefulness, and ease of use. The theoretical implications of this study's results within the framework of innovation diffusion theory are twofold. First, they confirm the role of perceived trialability in shaping behavioral intention, aligning with the theory's propositions. Second, the findings provide insights into the influence of other factors, such as perceived usefulness, ease of use, and trust, highlighting the need for a more comprehensive understanding of technology adoption. The study also raises questions about the mediating or moderating role of contextual factors, suggesting avenues for further investigation in the field of innovation diffusion theory.

This study focused on examining the factors that influence undergraduate students' intention to use chatbots for educational purposes within the framework of innovation diffusion theory. By analyzing these factors and their relationships with behavioral intention, the study offers valuable insights while also highlighting areas that require further investigation. One of the key findings of the study is the confirmation of several aspects of innovation diffusion theory. For instance, the study validated the significance of perceived relative advantages. It revealed that students who recognized the benefits of using chatbots for academic purposes expressed a strong intention to adopt them. Similarly, the research supported the role of compatibility, indicating that students who perceived a sense of compatibility with chatbot technology were more likely to adopt it. Additionally, the study confirmed the importance of trialability, as students who had the opportunity to try out chatbots were more inclined to use them. Lastly, the research validated the association between perceived usefulness and perceived ease of use and impact of perceived trust, showing that students with a higher level of trust in chatbot technology demonstrated a greater intention to adopt it. However, the study also provided insights that deviated from traditional expectations within an expanded innovation diffusion theory. The findings did not support the direct relationships between perceived usefulness, perceived ease of use, and behavioral intention. This suggests that in the context of chatbot adoption, these factors may not have as significant an influence as initially hypothesized. It implies that there may be additional factors or dynamics at play that require further exploration. The implications of this study hold practical significance for undergraduate students in higher education. The findings highlight the importance of creating awareness among students about the benefits of using chatbots and how they can be compatible with their learning practices. It is crucial to provide students with opportunities to trial and explore chatbot technology, allowing them to experience its potential benefits firsthand. Building trust in chatbot technology is also essential, as it positively influences students’ intention to adopt it for academic purposes. In conclusion, this study enhances our understanding of the factors influencing undergraduate students’ intention to use chatbots in higher education. The findings provide practical implications for students, emphasizing the importance of awareness, trialability, and trust. Additionally, the study contributes to the theoretical understanding of innovation diffusion theory. Future research can build upon these findings to gain deeper insights into the complexities of chatbot adoption and further enhance the integration of AI tools in educational settings.

This study makes substantial contributions to the existing literature in several key areas. Firstly, it uniquely focuses on the adoption of AI chatbots within an African context, specifically within the educational landscape of Lesotho. This emphasis adds a distinctive layer to the literature, offering insights into how cultural and contextual factors influence students' perceptions and intentions regarding the use of chatbots in higher education. Secondly, the study enriches the theoretical framework by incorporating cultural dimensions relevant to the African/Lesotho context. This enhancement enables a more nuanced exploration of the factors that shape students' attitudes and behaviors concerning AI technology. By integrating cultural factors, values, and beliefs into the theoretical framework, the study contributes to a deeper understanding of the intricate interplay between cultural context and technology adoption. Moreover, the research provides a comprehensive understanding of chatbot acceptance in African higher education, surpassing generic models. The study delves into specific cultural and contextual factors that influence the acceptance patterns of AI chatbots among students in Lesotho, offering valuable insights for researchers, practitioners, and policymakers. Additionally, the study fills a noteworthy research gap by specifically addressing AI adoption in Lesotho, contributing to the limited body of literature on technology adoption within African higher education institutions. This geographical focus provides a distinct contribution, shedding light on the unique challenges and opportunities associated with AI integration in the African educational landscape.

This study acknowledges several limitations that need to be addressed. Firstly, the findings are based on a specific sample of undergraduate students from a particular institution, so caution should be exercised when generalizing the results to other student populations or educational contexts. To ensure broader applicability, future research should include a more diverse range of participants. This means involving students from different universities, different academic programs, and diverse cultural backgrounds. By doing so, we can gain a better understanding of how these factors may influence chatbot adoption in various educational settings. Secondly, the cross-sectional design used in this study makes it challenging to establish causal relationships between variables. Since the data were collected at a single point in time, it is difficult to determine whether the factors examined actually influence students' intention to use chatbots or if there are other underlying factors at play. To overcome this limitation, future research should employ longitudinal studies that track students' intentions and actual usage of chatbots over an extended period. By observing changes in intentions and usage patterns over time, we can better understand the causal relationships and how they may evolve. Another limitation is the reliance on self-report measures, which may be subject to response biases and may not accurately capture students' true behaviors and intentions. To enhance the validity of the findings, future studies could consider incorporating objective measures or behavioral observations alongside self-report data. For example, researchers could collect data on the actual usage of chatbots by tracking students' interactions with the technology or analyzing system logs. This would provide more reliable and objective data on students' behaviors and intentions. Moreover, the study solely relied on survey data for data collection, which may limit the depth of understanding of students' experiences and perceptions. To gain a more comprehensive understanding of the factors influencing chatbot adoption, future studies could incorporate qualitative interviews or observations in addition to surveys. Qualitative research methods can delve deeper into students' experiences, attitudes, and perceptions regarding chatbot adoption. By conducting interviews or observations, researchers can explore specific challenges, concerns, and motivations that influence students' decision-making processes and shed light on the underlying reasons behind their intentions. Furthermore, experimental designs can be utilized to examine the causal relationships between the identified factors and students' intention to use chatbots. By manipulating variables in controlled settings, researchers can establish a cause-and-effect relationship and provide stronger evidence of the factors' impact on adoption. Finally, investigating the influence of contextual factors, such as educational disciplines or specific learning tasks, on chatbot adoption would provide a nuanced understanding of the technology's applicability and effectiveness in different academic contexts. By examining how these factors interact with students' intentions and behaviors, researchers can identify potential barriers or facilitators to chatbot adoption and tailor the implementation strategies accordingly.