Supervisor: pantelis papadopoulos
Collaborative learning is a widely used approach to engage students in learning activities. However, balanced engagement, meaningful contributions, and shared goals are necessary for a productive collaboration. In that sense, transactivity, or building upon a previous contribution of a learning partner, is important for the co-construction of knowledge during collaborative learning (Weinberger & Fischer, 2006).
Supporting transactive dialogue is one of the main goals for collaborative conversational agents (CCAs) based on artificial intelligence (AI). However, the there is no consensus on how to define and measure transactivity as several models can be found in the literature.
Following the Vogel and Weinberger model for transactivity (2023) which operationalizes the roles of novelty and reference in a dialogue, this internship will focus on prompting ChatGPT for the automatic evaluation of transactivity in student dialogues in collaborative problem-solving activities.
METHOD
The internship will use both qualitative and quantitative methods. Already recorded dialogues will be analyzed using Atlas.ti. The human-driven coding will be used as the baseline for the analysis of ChatGPT performance. Different prompting techniques will may be employed and comparison of their output will be analyzed with inferential statistics.
REFERENCES
Vogel, F., Weinberger, A., Hong, D., Wang, T., Glazewski, K., Hmelo-Silver, C. E., Uttamchandani, S., Mott, B., Lester, J., Oshima, J., Oshima, R., Yamashita, S., Lu, J., Brandl, L., Richters, C., Stadler, M., Fischer, F., Radkowitsch, A., Schmidmaier, R., Fischer, M. R., Rejon, M. A., & Noroozi, O. (2023). Transactivity and knowledge co-construction in collaborative problem solving. In Damșa, C., Borge, M., Koh, E., & Worsley, M. (Eds.), Proceedings of the 16th International Conference on Computer-Supported Collaborative Learning - CSCL 2023 (pp. 337-346). International Society of the Learning Sciences. https://repository.isls.org/bitstream/1/9227/1/CSCL2023_337-346.pdf
Weinberger, A., & Fischer, F. (2006). A framework to analyze argumentative knowledge construction in computer-supported collaborative learning. Computers & education, 46(1), 71-95. https://doi.org/10.1016/j.compedu.2005.04.003
Prompt Engineering Guide: https://www.promptingguide.ai/techniques/tot