Supporting Students to Identify a Research Project Topic

Delays selecting a feasible research topic cause students anxiety and distress. Technologies like ChatGPT pose challenges to higher education delivery and assessment, but also create opportunities to address individual learning needs. This study was to develop, implement, and evaluate a GenAI chatbot to support students to identify a research topic. A 4-stage mixed-methods study guided by the active digital learning pedagogy: collation and synthesis of 252 former students’ project topics, followed by development, implementation, and evaluation of a GenAI topic-bank chatbot. An online survey including the 11-item Bot Usability Scale and a 9-item purposefully developed survey based on active digital learning pedagogy was used to collect students’ perceptions of chatbot usability and learning experiences. Students’ overall learning experiences were explored using 2 open-ended questions. All 23 students, enrolled into the capstone project unit of semester 1 2025, utilized the chatbot, across 35 conversations and 233 searches. Each student initiated 1 to 3 conversations and 3 to 23 searches. Average bot usability scale score was 3.57/5 and learning evaluation was 3.95/5. Narratively, students shared mixed experiences and perceptions. The chatbot generally facilitated students in identifying project topics and improved their learning experience. Built on customized knowledge, the chatbot has potential application to other postgraduate courses’ students.