Design and Development of AI-Based Classroom Activity Dashboards
Abstract: Advances in neural networks (i.e., a form of artificial intelligence) have the potential to make the process of analyzing classroom videos less time-consuming and more efficient. In this qualitative study, we collaborated with teachers and instructional experts to develop automated personalized dashboards to inform instructional decisions on classroom activities. We leveraged the Adaptive System Framework to inform the design of automated dashboards. Although participants differed in opinion on what specific types of automated data should be presented, this study provides initial evidence for using neural network data to automate teacher dashboards. We highlight the importance of including non-technical experts (i.e., teachers, instructional coaches, and instructional designers) in the design and development of dynamic classroom activity dashboards.
Presider: Stepohanie Robles