Students closed a higher relative number of knowledge gaps when a red notification badge nudged them than when it was turned off¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1–r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In a within-student analysis, students' relative number of closed knowledge gaps was significantly higher when the notification badge was turned on than when it was turned off (b = 0.43, p < .001). → Spitzer 2025 q2 i? In a between-student analysis, students who used the ITS only when the badge was on closed about 25% of their knowledge gaps versus about 5% for students who used it only when the badge was off (b = 0.30, p < .001). → Spitzer 2025 (2)

Evidence¶

Spitzer 2025¶

Spitzer, M. W. H., Moeller, K., Richter, E., Frede, L., & Bardach, L. (2025). A red button does it all: A red button nudges students to close their knowledge gaps in an intelligent tutoring system. Learning and Instruction 99 (2025) 102181. https://doi.org/10.1016/j.learninstruc.2025.102181

q2 · i? · causal · r1

Within-student hierarchical linear regression on 28,088 students who used bettermarks in both periods, with students nested in classes. With the badge on, students closed about 33% of knowledge gaps versus about 5% with it off; the difference was "statistically significant" (b = 0.43, p < .001). No standardized effect size is printed for this contrast.

"the difference in closing knowledge gaps when the notification badges were turned off vs. turned on was statistically significant for both the within-student analysis (b = 0.43, SE <.01, t = 122.27, p < .001) and the between-student analysis (b = 0.30, SE <.01, t = 39.51, p < .001)"

Spitzer 2025 (2)¶

Spitzer, M. W. H., Moeller, K., Richter, E., Frede, L., & Bardach, L. (2025). A red button does it all: A red button nudges students to close their knowledge gaps in an intelligent tutoring system. Learning and Instruction 99 (2025) 102181. https://doi.org/10.1016/j.learninstruc.2025.102181

q2 · i? · causal · r2

Between-student hierarchical linear regression comparing 68,784 badge-on students with 8,209 badge-off students, with a random intercept for classes. Badge-on students closed about 25% of gaps versus about 5% for badge-off students (b = 0.30, p < .001). No standardized effect size is printed for this contrast.

"with the notification badge turned on, students' relative number of closed knowledge gaps was about 33 % for the within-student and 25 % for the between-student analysis"

Discussion¶

Related Claims¶

• In a hierarchical regression, class-level knowledge gap closing had the largest beta weight of all considered predictors of performance — related
• Closing knowledge gaps showed the highest association with performance among all considered variables (partial r = 0.17) — related
• Closing knowledge gaps between a baseline and an end period predicted higher performance in the end period — related