<![CDATA[The development of LLMs like ChatGPT revolutionizes the world of higher education through providing customized explanations, interaction in problem-solving activities, and feedback instantly. However, the successful use of ChatGPT for learning activities is not only dependent on the presence of an AI system but on the proficiency of the student in creating effective prompts as well. The current study investigated the impact of Prompt Engineering Ability on ChatGPT Output Quality and Learning Effectiveness, as well as the subsequent associations among Performance Expectancy, Effort Expectancy, and Usage Intensity in accordance with the Unified Theory of Acceptance and Use of Technology (UTAUT). The data collection took place in the Probability Theory course and involved 385 undergraduate students. PLS-SEM analysis was performed in SmartPLS 4. The results indicated that Prompt Engineering Ability positively affected ChatGPT Output Quality (β = 0.599), which, in turn, enhanced Learning Effectiveness (β = 0.607). Both Prompt Engineering Ability and ChatGPT Output Quality positively affected Performance Expectancy (β = 0.372 and 0.367), which then influenced Effort Expectancy (β = 0.602) and led to more Usage Intensity (β = 0.578). These results combine the concepts of Prompt Engineering Ability, ChatGPT Output Quality, and UTAUT in a framework that is beneficial for Artificial Intelligence in Education (AIED) research in higher education. ]]>
Copyrights © 2026
