Data Inputs and Contexts for ChatGPT in Education: A Review of Empirical Studies
Wenting Sun, Humboldt-Universität zu Berlin (Germany)
Jiangyue Liu, Suzhou University (China)
Xiaoling Wang, Zhejiang Normal University (China)
Abstract
Despite some education stakeholders' attempts to ban ChatGPT, many support its meaningful and effettive use in teaching and learning [1,2,3]. Lo et al. [4] argue that the misuse of this tool often results from a lack of knowledge and understanding of its proper use and encourage develop clear guidelines for the acceptable use of AI tools like ChatGPT. Recently, significant efforts have been made to design appropriate prompts. However, the data and context for their use in education remain understudied. This review, following the Preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidance, selected 102 empirical articles with detailed prompts when use ChatGPT from Web of science and Scopus. Qualitative method was utilized to extract information about educational level, knowledge domain, research method, ChatGPT affordance in education, data source (embedded in prompting), data collection method. Teaching and learning experiences were summarized from these articles. It was found that higher education is the main educational levels adapted ChatGPT. Language learning and education in humanity while computer science, chemistry, mathematics, and medicine in STEM disciplines gained more attention. Half of the included articles chose qualitative method. When using and analysing ChatGPT, data collected from or related to course level was the main source for embedding context or task information into prompting, and data generated by ChatGPT was the main data source for further analysis of the ChatGPT outputs based on some prompts (rather than users’ perceptions and performance impacted by ChatGPT). Moreover, ChatGPT provides more diverse affordances in two course phases, during the courses and after the courses respectively. Some teaching strategies and learning strategies were summarized to enhance students' critical or reasoning thinking.
Keywords: ChatGPT in Classroom, Affordances, Prompting, AI Tools in Teaching
REFERENCES
[1] Mai D. T. T., Da C. V., Hanh N. V., “The use of ChatGPT in teaching and learning: a systematic review through SWOT analysis approach”, Frontiers in Education, 2024, 9, 1328769.
[2] Kooli C., “Chatbots in education and research: A critical examination of ethical implications and solutions”, Sustainability, 2023, 15(7), 5614.
[3] Ali J. K. M., Shamsan M. A. A., Hezam T. A., Mohammed A. A., “Impact of ChatGPT on learning motivation: teachers and students' voices”, Journal of English Studies in Arabia Felix, 2023, 2(1), 41-49.
[4] Lo C. K., Hew K. F., Jong M. S. Y., “The influence of ChatGPT on student engagement: A systematic review and future research agenda”, Computers & Education, 2024, 105100.