Major education and training challenges are plaguing the United States in preparing the next generation of the future workforce to meet the demands of the 21st Century. Several calls have been released to improve education programs to ensure learners are acquiring 21st century knowledge, skills, and abilities (KSAs). As we embark on the digital and automation ages of the 21st century, it is essential that we move away from traditional education programs that define and measure KSAs as static constructs (e.g., standardized assessments) with little consideration of the actual real-time deployment of these processes, missing critical information on the degree to which learners are acquiring and applying 21st century KSAs. The objective of this dissertation is to use 1 book chapter and 2 journal articles to illustrate the value in leveraging emerging technologies and multimodal trace data to define and measure scientific thinking, reflection, and self-regulated learning--core 21st century skills, across contexts, domains, tasks, and populations (e.g., medical versus undergraduates versus middle-school students). Chapters 2-4 of this dissertation provide evidence of ways to leverage multimodal trace data guided by theoretical perspectives in cognitive and learning sciences, with a special focus in self-regulated learning, to assess the extent to which learners engaged in scientific thinking, reflection, and self-regulated learning during learning activities with emerging technologies. Overall, results from these chapters illustrate that it is necessary to utilize methods that capture learning processes as they unfold during learning activities that are guided by theoretical perspectives in self-regulated learning. Findings from this research hold significant broader impacts for addressing the education and training challenges in the United States by collecting multimodal trace data over the course of learning to not only detect and identify how learners are developing KSAs such as scientific thinking, reflection, and self-regulated learning, but where these data could be fed into an intelligent and adaptive system to repurpose it back to trainers, teachers, instructors, and learners for just-in-time interventions and individualized feedback. The intellectual merit of this dissertation focuses predominantly on the importance of utilizing rich streams of multimodal trace data that are mapped onto different theoretical perspectives on how humans self-regulate across tasks like clinical reasoning, scientific thinking, and reflection with emerging technologies such as a game-based learning environment called Crystal Island. Discussion is incorporated around ways to leverage multimodal trace data on undergraduate, middle-school, and medical student populations across a range of tasks including learning about microbiology to problem solving with a game-based learning environment called Crystal Island and clinically reasoning about diagnoses across emerging technologies.
If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu.
Doctor of Philosophy (Ph.D.)
College of Community Innovation and Education
Learning Sciences and Educational Research
Education; Learning Sciences
Length of Campus-only Access
Doctoral Dissertation (Open Access)
Cloude, Elizabeth, "Leveraging Multimodal Learning Analytics to Understand How Humans Learn with Emerging Technologies" (2021). Electronic Theses and Dissertations, 2020-. 839.