What Does a Digital Trace Mean? Validating Inferences in Learning Analytics

• In Event: Measuring SRL Processes in STEM College Courses

Abstract

Digital learning environments provide researchers information about learners’ self-regulated learning (SRL) processes, but the validity of inferences drawn from digital events and sequences must be established by aligning digital events to a trustworthy channel of SRL data. We recruited 49 participants from a college algebra course who thought aloud while completing a lesson we adapted for a laboratory study. We aligned learners' verbalizations to their digital traces and sequences of digital events to examine how consistently digital events and sequences co-occurred with verbal descriptions of SRL processes. Many digital traces and several sequences tracked reliably with verbalized SRL. Learners’ characteristics moderated verbal and digital trace co-occurrence. Validating digital traces demonstrated predictive validity for achievement.

Authors

• Matthew L. Bernacki, University of North Carolina - Chapel Hill

• Linyu Yu, University of North Carolina - Chapel Hill

• Shelbi Laura Kuhlmann, University of Memphis

• Robert D Plumley, University of North Carolina - Chapel Hill

• Jeff A. Greene, University of North Carolina - Chapel Hill

• Peter F. Halpin, University of North Carolina - Chapel Hill

• Rebekah F. Duke, University of North Carolina - Chapel Hill

• Rebekah Freed, University of Washington

• Christina Hollander-Blackmon, University of North Carolina - Chapel Hill

• Miranda D. Thomas, University of North Carolina - Chapel Hill