Systematic Review of Process Data Analytics in K-12 Education

• In Event: Reimagining Assessment as a Learning Process: Student Voice, Classroom Practice, and Data Analytics

Abstract

This systematic review examines how process data analytics has been used to understand K–12 students' cognition, behavior, and emotion in computer-based learning environments. Analyzing 70 peer-reviewed studies, we identify the aspects of learning examined, data types, analytical methods, and domain-specific insights. Most studies focus on cognition and behavior, while emotional aspects remain underexplored. Action- and time-related data dominate in modeling problem-solving, self-regulation, and engagement, whereas multimodal and biometric data are emerging for emotion analysis. Analytical approaches vary by domain: time-series and latent models for cognition, clustering for behavior, and variance analysis for emotion. The review emphasizes complex learning metrics and the need for multidimensional methods to support adaptive learning systems.

Authors

• Yaxuan Yang, University of Georgia

• Ruiping Huang, University of Illinois at Chicago

• Yue Yin, University of illinois at Chicago

• Xiaoming Zhai, University of Georgia