As support chatbots become more prevalent in educational environments, particularly those powered by Large Language Models (LLMs), understanding the impact of different dialogue strategies is increasingly important. Chatbots can help students with personalized learning and guidance, but the effectiveness of their approach depends on how they engage learners. Two main conversational strategies are:

• Tutoring-based content delivery: A directive approach providing structured guidance and feedback.
• Socratic-based questioning: An open-ended, questioning method aimed at fostering critical thinking and self-reflection.

Despite theoretical support for both strategies, little empirical research has compared their specific effects on student outcomes such as engagement, goal attainment, and learning retention.

• Bachelor’s Thesis: Conduct a systematic literature review to explore and evaluate the differential effects of tutoring and Socratic coaching strategies in LLM-driven chatbots. This will involve:

  • Reviewing theoretical frameworks for tutoring and coaching methods in educational settings.
  • Synthesizing empirical findings on their impact on student engagement, cognitive load, and learning outcomes.
  • Identifying research gaps and opportunities for future work.
  • May include light-weight testing of insights (depending on preexisting skill set).

• Master’s Thesis: In addition to the literature review, a Master’s thesis will include an experiment to test the insights from the review. The student will:

  • Design and conduct an experiment comparing the two strategies in a controlled environment.
  • Measure outcomes such as engagement, goal achievement, and retention. 
  • Analyze the results to provide evidence-based recommendations on chatbot dialogue strategies for improved student learning.

• Interest in educational technology, AI-driven chatbots, and learning sciences.
• Strong research and analytical skills for conducting a comprehensive literature review (BA).
• For Master’s applicants: Ability to design and implement an experiment, with experience in data analysis being a plus.
• Proficiency in scientific writing in English.

• Supervision and feedback from experts in AI and educational technologies.
• Flexible timelines and a supportive research environment.

• Your CV.
• Academic transcripts.
• A short motivation letter (max. 200 words) indicating your interest and whether you are applying for a Bachelor's or Master’s thesis.