TY - JOUR

T1 - Research trends in multimodal learning analytics

T2 - A systematic mapping study

AU - Ouhaichi, Hamza

AU - Spikol, Daniel

AU - Vogel, Bahtijar

N1 - Funding Information: We want to thank the authors and researchers whose work we reviewed and analyzed as part of this literature review. Without their contributions, this research would not have been possible. We would also like to thank the organizations that provided access to data and resources for this research. We are grateful for their commitment to open data and their willingness to share their data with us. Finally, we thank our colleagues and peers for their feedback and support throughout the research process. Their insights and guidance have been invaluable to this work. We also want to disclose any conflicts of interest that may have influenced this research. Again, we have no conflicts of interest to report. We also want to emphasize that this research was conducted following ethical principles, including respect for the autonomy, confidentiality, and privacy of the authors and researchers whose work we reviewed. Furthermore, all data were collected and analyzed by relevant laws and regulations. Publisher Copyright: © 2023 The Authors

PY - 2023

Y1 - 2023

N2 - Understanding and improving education are critical goals of learning analytics. However, learning is not always mediated or aided by a digital system that can capture digital traces. Learning in such environments can be studied by recording, processing, and analyzing different signals, including video and audio, so that traces of actors’ actions and interactions are captured. Multimodal Learning Analytics refers to analyzing these signals through the use and integration of these multiple modes. However, a need exists to evaluate how research is conducted in the emerging field of multimodal learning analytics to aid and evaluate how these systems work. With the growth of multimodal learning analytics, research trends and technologies are needed to support its development. We conducted a systematic mapping study based on established systematic literature practices to identify multimodal learning analytics research types, methodologies, and trending research themes. Most mapped papers presented different solutions and used evaluation-based research methods to demonstrate an increasing interest in multimodal learning analytics technologies. In addition, we identified 14 topics under four themes––learning context, learning process, systems and modality, and technologies––that can contribute to the growth of multimodal learning analytics.

AB - Understanding and improving education are critical goals of learning analytics. However, learning is not always mediated or aided by a digital system that can capture digital traces. Learning in such environments can be studied by recording, processing, and analyzing different signals, including video and audio, so that traces of actors’ actions and interactions are captured. Multimodal Learning Analytics refers to analyzing these signals through the use and integration of these multiple modes. However, a need exists to evaluate how research is conducted in the emerging field of multimodal learning analytics to aid and evaluate how these systems work. With the growth of multimodal learning analytics, research trends and technologies are needed to support its development. We conducted a systematic mapping study based on established systematic literature practices to identify multimodal learning analytics research types, methodologies, and trending research themes. Most mapped papers presented different solutions and used evaluation-based research methods to demonstrate an increasing interest in multimodal learning analytics technologies. In addition, we identified 14 topics under four themes––learning context, learning process, systems and modality, and technologies––that can contribute to the growth of multimodal learning analytics.

KW - Artificial intelligence

KW - Learning technologies

KW - Mapping study

KW - Multimodal learning analytics

U2 - 10.1016/j.caeai.2023.100136

DO - 10.1016/j.caeai.2023.100136

M3 - Review

AN - SCOPUS:85151456109

VL - 4

JO - Computers and Education: Artificial Intelligence

JF - Computers and Education: Artificial Intelligence

SN - 2666-920X

M1 - 100136

ER -