Unobtrusive observations of learning in digital environments examining behavior, cognition, emotion, metacognition and social processes using learning analytics

Includes bibliographical references and index.

Learning Processes — Unobtrusive Observations of Learning Processes — Section Overview — A Review of Measurements and Techniques to Study Emotion Dynamics in Learning — The Features of Emotion Dynamics — Emotional Variability — Emotional Instability — Emotional Inertia — Emotional Cross-lags — Emotional Patterns — The Measurements of Emotion Dynamics — Experience Sampling Method — Emote-Aloud — Facial Expressions — Vocal Expressions — Language and Discourse — Physiological Sensors — The Techniques for Analyzing Emotion Dynamics — Conventional Statistical Methods — Entropy Analysis — Growth Curve Modeling — Time Series Analysis — Network Analysis — Recurrence Quantification Analysis — Sequential Pattern Mining — The Challenges of Studying Emotion Dynamics in Learning — Deciding What to Measure About Emotion Dynamics — Deciding How to Analyze Emotion Dynamics — Addressing Individual and Developmental Differences — Differentiating Between Short-Term and Long-Term Emotion Dynamics — Concluding Remarks and Directions for Future Research — References — Applying Log Data Analytics to Measure Problem Solving in Simulation-Based Learning Environments — Introduction — Background — Methods — Experiment — Experiment — Log Data Processing — Results — Problem-Solving Outcomes as Measured by Solution Quality — Problem-Solving Processes as Captured by Features Extracted from Log Data — Pause as a Generalizable Indicator of Deliberate Problem Solving — How Log Data-Based Features Were Associated with Specific Problem-Solving Practices — Discussion — Limitations.

This book integrates foundational ideas from psychology, immersive digital learning environments supported by theories and methods of the learning sciences, particularly in pursuit of questions of cognition, behavior and emotion factors in digital learning experiences. New and emerging foundations of theory and analysis based on observation of digital traces are enhanced by data science, particularly machine learning, with extensions to deep learning, natural language processing and artificial intelligence brought into service to better understand higher-order thinking capacities such as self-regulation, collaborative problem-solving and social construction of knowledge. As a result, this edited volume presents a collection of indicators or measurements focusing on learning processes and related behavior, (meta-)cognition, emotion and motivation, as well as social processes. In addition, each section of the book includes an invited commentary from a related field, such as educational psychology, cognitive science, learning science, etc.