Assessing Computational Thinking through Learners’ Embodied Interactions and the Lens of Complex Dynamical Systems Theory

The purpose of this dissertation is to explore how different methodological possibilities can support the analysis of computational thinking (CT) from an embodied perspective. To that end, this dissertation first presents an approach to multimodal transcription using a methodological framework that aligns with the study of CT as an embodied phenomenon. The framework, which incorporates a social semiotic approach to multimodality, then is applied to the same multimodal transcript using three different analytical approaches: (1) a grounded approach, (2) a process of transduction with a priori coding and thematic analysis, and (3) an artificial intelligence (AI) pattern recognition approach. Each approach is presented as a case, allowing each to be evaluated within as well as across approaches. The criteria for evaluation include two major characteristics of embodied cognition and complex dynamical systems: self-organization and emergence (Gallagher & Appenzeller, 1999; Richardson & Chemero, 2017). This research contributes meaningfully to current and ongoing questions about embodied cognition in that it identifies the strengths and weaknesses associated with different approaches to the analysis of CT from a social semiotic perspective of multimodality.