Abstract:

This contribution reports on student acceptance of EQAssistant, a conversational chatbot developed at the University of the Balearic Islands to support autonomous learning in Chemical Engineering subjects. The ChatBot structures problem-solving into a conversational flow in which students receive 8 questions (5 multiple-choice and 3 open-ended). The current implementation addresses a representative course problem (mass and energy balances in a heat exchanger), but the design is generic and can be easily adapted to other topics. The flow comprises 47 steps where each student response triggers tailored feedback; when mistakes occur, the ChatBot provides immediate explanations. In addition, its natural language processing functionality allows it to recognise different student inputs, increasing flexibility and personalisation. Being available online at any time, EQAssistant offers an “always-on” resource to support autonomous study.

The tool was offered to two cohorts: second-year Chemistry undergraduates enrolled in a compulsory Chemical Engineering course (N = 12) and Industrial Engineering Master’s students taking an Introduction to Engineering course (N = 11). After working with EQAssistant, students completed an online satisfaction questionnaire.

Results show high perceived usefulness for autonomous study at home in both groups (83% of undergraduates and 91% of Master’s students responded “yes”) and a strong intention to reuse the tool if similar exercises were proposed (91% of undergraduates and 73% of Master’s students indicated “probable” or “very probable”). Perceived innovativeness was moderately high, with mean ratings of 3.7/5 and 3.5/5 for the undergraduate and Master’s cohorts, respectively. Open-ended comments in both groups valued the step-by-step guidance and immediate feedback but highlighted a strong desire to ask additional, unstructured questions.

This study compares perceptions across both cohorts and outlines practical implications for integrating conversational agents to support autonomous learning.

Acknowledgements
This work was supported by the Universitat de les Illes Balears within the 2025–2027 Call for Teaching Innovation and Quality Improvement Projects (project PID252744 “Integración de ChatBots y ChatGPT como Herramientas de Aprendizaje Autónomo y Pensamiento Crítico en Ingeniería Química”), with funding from the Vice-Rectorate for Teaching and Research Staff, the Vice-Rectorate for Strategic Planning, Internationalisation and Cooperation, and the Institute for Educational Research and Innovation (IRIE, UIB–GOIB).

Keywords:

Chatbot, chemistry, engineering, autonomous learning, undergraduate, master.