Workload peaks and slope variability correlate with perceived learning attainment, with LMS dynamics showing no significant correlations¶

Claim · All claims
Evidence · 1 study (3 entries) · associational r1 · q3 · i3 large
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i3 Course Average Learning is negatively correlated with the number of Total Peaks (r=-0.57) and Low-Intensity Peaks (r=-0.61). → Paul V. Sargent 2026 q3 i3 The percentage of students reporting learning much more than expected correlates negatively with Max Positive Slope (r=-0.70), Positive Slope SD (r=-0.70), and Course Time-on-Task SD (r=-0.69). → Paul V. Sargent 2026 (2) q3 i? No LMS-derived workload distribution metric showed significant correlations with learning outcomes. → Paul V. Sargent 2026 (3)

Evidence¶

Paul V. Sargent 2026¶

Paul V. Sargent, Isabel Hilliger, Jorge A. Baier. (2026). Beyond Time on Task: A Novel Analytical Framework for Assessing Student Workload and Its Relationship with Learning. Journal of Learning Analytics, 13(1), 110–129. https://doi.org/10.18608/jla.2026.8725

q3 · i3 · associational · r1

Course-level correlation analysis of WTS workload distribution metrics against course evaluation survey learning attainment across 14 courses. Total Peaks correlated at "r=−0.57" and Low-Intensity Peaks at "r=−0.61" with Course Average Learning.

"First, we found that Course Average Learning was negatively correlated with the number of Total Peaks (r=−0.57 )—regardless of their type—particularly with the presence of Low-Intensity Peaks ( r=−0.61 )."

Paul V. Sargent 2026 (2)¶

Paul V. Sargent, Isabel Hilliger, Jorge A. Baier. (2026). Beyond Time on Task: A Novel Analytical Framework for Assessing Student Workload and Its Relationship with Learning. Journal of Learning Analytics, 13(1), 110–129. https://doi.org/10.18608/jla.2026.8725

q3 · i3 · associational · r1

Course-level correlations between WTS dynamics metrics and the percentage of students reporting learning much more than expected (MMTE) on the course evaluation survey, all negative and statistically significant.

"Conversely, the percentage of students that reported learning MMTE showed negative correlations with Max Positive Slope (r=−0.70 ), Positive Slope SD (r=−0.70 ), and Course Time-on-Task SD (r=−0.69 )."

Paul V. Sargent 2026 (3)¶

Paul V. Sargent, Isabel Hilliger, Jorge A. Baier. (2026). Beyond Time on Task: A Novel Analytical Framework for Assessing Student Workload and Its Relationship with Learning. Journal of Learning Analytics, 13(1), 110–129. https://doi.org/10.18608/jla.2026.8725

q3 · i? · associational · r1

Null result from the same course-level analysis: LMS-derived distribution metrics showed no significant correlations with Course Average Learning, despite WTS counterparts being significant. No effect sizes are interpretable as magnitudes here.

"neither LMS Time-on-Task SD (r=−0.16 , p=0.59 ), LMS Overall Slope SD (r=0.02 , p=0.95 ), nor LMS Max Positive Slope (r=−0.05 , p=0.87 ) showed significant correlations with Course Average Learning"

Discussion¶

Related Claims¶

• Workload variability and slope metrics correlate strongly with perceived course difficulty — related
• Self-reported weekly time-on-task is positively correlated with LMS-based time-on-task and activity measures in engineering courses — related
• Quizzes are the only LMS activity type negatively correlated with reported time-on-task and perceived difficulty — related
• Courses with higher average time-on-task exhibit greater variability in their weekly workload evolution — related