The mixed algorithm produced groups differing significantly in marker-feature distributions, with shared (overlapping) yellow markers significantly lower than unique yellow markers¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2 · i3 large
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly

Subclaims¶

q2 i3 A one-way ANOVA on the four group marker features showed a statistically significant difference among groups (F(3, 36) = 7.0755, p = .0007), with η² = 0.371. → Liang 2025 q2 i? Shared Yellow Markers were significantly lower than Unique Yellow Markers (p = .002), while the Unique Red vs Shared Red comparison showed no significant difference (p = .149). → Liang 2025 (2)

Evidence¶

Liang 2025¶

Liang, C., Toyokawa, Y., & Ogata, H. (2025). Optimizing group formation with a mixed genetic algorithm: an empirical study in active reading using marker data. Intern. J. Comput.-Support. Collab. Learn. https://doi.org/10.1007/s11412-025-09452-9

q2 · i3 · design · r2

Descriptive analysis of the 10 algorithm-formed groups' marker features (unique/shared red and yellow markers) followed by one-way ANOVA after normality and variance checks. The test reached significance with η² = 0.371 ("37.1% of the variance explained by marker features").

"The ANOV A results indicated a statistically significant difference among the groups ( F(3, 36) = 7.0755, p = .0007). The effect size, measured as Eta Squared (η2), was 0.371, indicating a medium-to-large effect size, with 37.1% of the variance explained by marker features."

Liang 2025 (2)¶

Liang, C., Toyokawa, Y., & Ogata, H. (2025). Optimizing group formation with a mixed genetic algorithm: an empirical study in active reading using marker data. Intern. J. Comput.-Support. Collab. Learn. https://doi.org/10.1007/s11412-025-09452-9

q2 · i? · design · r2

Tukey's HSD pairwise comparisons of marker features. Shared Yellow was significantly lower than Unique Yellow (p = .002); Unique Red vs Shared Red was non-significant (p = .149), as were the remaining red-yellow pairs.

"The Shared Yellow Marker exhibited significantly lower values compared to the Unique Yellow Marker ( p = .002 <. 05). When comparing the Unique Red Marker and the Shared Red Marker , although there were more unique markers for the red marker, no significant difference was found ( p = .149 >. 05)."

Discussion¶

Related Claims¶

• Process mining of two fitness-typical groups suggests the group with more heterogeneous yellow markers and shared red markers worked more efficiently while both achieved the same summary outcome
• In a high school EFL active-reading setting, mixed genetic algorithm group formation with Wordcloud awareness produced significantly greater vocabulary learning gains than random grouping
• A text-mining group and representation tool scaffolds students' learning processes and improves outcomes under specific conditions — a broader claim this one bears on