Title:
Leveraging computationally generated descriptions of audio features to enrich qualitative examinations of sustained uncertainty (Download PDF)

Citation:
Krist, C., Dyer, E., Rosenberg, J., Palaguachi, C., & Cox, E. (2023, June). Leveraging computationally generated descriptions of audio features to enrich qualitative examinations of sustained uncertainty. In General Proceedings of the 16th International Conference on Computer-Supported Collaborative Learning 2023.

Abstract:
Prosodic features of speech, such as pitch and loudness, are important aspects of the social dimensions of learning. In particular, these features are likely related to sustained disciplinary uncertainty in collaborative STEM learning contexts. We present a case conducting an exploratory, descriptive analysis of sustained uncertainty in groupwork in a secondary mathematics lesson integrating computational and qualitative methods with audiovisual data. Results of computational audio feature extraction of loudness and pitch, combined with a transcript, were used to identify potential patterns between laughter and uncertainty.

Title:
Audio Analysis of Teacher Interactions with Small Groups in  Classrooms (Download PDF)

Citation:
Palaguachi, C., Cox, E., & D’Angelo, C. (2022, June). Audio Analysis of Teacher Interactions with Small Groups in Classrooms. In General Proceedings of the 15th International Conference on Computer-Supported Collaborative Learning 2022.

Abstract:
This paper presents exploratory work in combining computational methods with qualitative approaches in order to better understand teacher interactions with small groups of students. Classroom audio data is typically difficult to work with, due to background noise and challenging acoustics, and needs customization of algorithm parameters when using audio processing tools for speech detection. This secondary data analysis study looked at patterns in small group discussions of students over multiple class sessions and multiple teachers, especially focusing on times when teachers interacted with the groups. This type of approach can augment and extend the capabilities of qualitative researchers, who could use these computationally-derived analytics and patterns to aid them in better understanding teacher/student interactions and collaborative learning.