Analyzing Student Team Dialogues To Guide The Design Of Active Learning Sessions

Engineering faculty are increasingly using active learning methods to improve learning in their classes. Many methods and their uses are described in the literature. These methods range from impromptu techniques such as “think-pair-share” up to strategies for structuring the entire course. The strength of these methods relies on generating student interactions marked with deep cognitive reasoning. Presumably, the greater the depth of reasoning in interactions, the greater the potential for learning. We define deep interactions as conversations that show thoughtful use of schema to organize information and/or the organization of information to create schema. In contrast, shallow interactions deal primarily with exchange of information. Active learning methods, when used properly, initiate deep student interactions. However, many teaching environments do not directly fit into the prescription of a well-researched method. Consequently, at times faculty must thoughtfully adapt these methods for their classes. However, in doing so there is no guarantee that deep interactions will ensue. Furthermore, faculty may also wish to diagnose whether their application of an active learning method is working as planned. One way to assess active learning is to assess the depth of the student interactions. These interactions may be assessed by recording, transcribing, and analyzing student dialogues. Our question is: What important design features for active learning sessions can be identified by the use of brief analyses of student dialogue? This case study examines the student dialogues in four sequential active learning sessions. In each session, a student team was recorded and their conversation transcribed. The transcription was reviewed and the observations were used to improve the design of the next session. After the conclusion of the sessions, the transcripts were examined for trends that emerged across multiple sessions. Three findings emerged: 1. Briefly coding transcripts by identify major themes and then coding along those themes surfaced substantial feedback to improve the design of the active sessions. The use of coding criteria, such as the three principles of learning, was used informally to interpret the content of the coding. The iterative use of transcript coding and session improvement created sessions with dialogues showing deeper interactions. 2. The student learning appeared to be tied to context. When the case supplied the context, the students used it to create schema. When the context was not supplied, the students created their own context to use. Consequently, cases that provide a rich context appear to better support the use of schemas related to the case. 3. The students seemed to intuitively identify the challenge in each session and apply their efforts to resolving it. This included challenges that were unintentionally introduced into the case. Including interdependencies within the case, that is information that cannot be resolved serially, appears to be one way to add challenge that leads to deeper interactions.