Theory and Intermediate-Level Knowledge in Multimodal Learning Analytics

Advances in sensing technologies and computational analyses have demonstrated the potential to help us understand learning processes which were either not-possible to be captured or “too complex” for traditional analytics (e.g., eye-tracking, temperature sensing, hear-rate sensing). This gave rise to the research space of Multimodal Learning Analytics (MMLA), which maintains Learning Analytics’ overarching goal of understanding and improving learning in all the different environments where it occurs, but also leverages advances in sensor technologies and computational analyses (e.g., sensor data processing, fusion, and analysis techniques). In this chapter, we introduce the reader to the field of MMLA and provide an overview of contemporary MMLA research. Moreover, we give an overview of theories used in MMLA and discuss the potential of MMLA to support the development of intermediate-level knowledge and how this knowledge, can be generated. To exemplify the importance and potential of intermediate-level knowledge in MMLA, we provide three examples on how knowledge that is more generative than an instantiation and yet not at the scope of generalized theory, manages to greatly advance MMLA research and practice.