Grade 5 students in a KBM environment significantly increased their Earth Science domain knowledge from pretest to post-test¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2–q3 · i1 small
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i1 Students' domain knowledge increased significantly following the KBM approach, with a small-to-moderate effect. → Dina Soliman 2026

Evidence¶

Dina Soliman 2026¶

Dina Soliman, Xueqi Feng, Marlene Scardamalia, Carol Chan. (2026). Knowledge building–modeling: Grade 5 students creating models, using discourse analytics, and advancing systems thinking. Intern. J. Comput.-Support. Collab. Learn. https://doi.org/10.1007/s11412-026-09493-8

q3 · i1 · design · r2

Paired-sample t-test on the same ten open-ended Earth Science items administered as pretest and post-test in the KBM unit. The article reports 'significantly higher domain test scores in the post-test' with 'Cohen's d = 0.37'. Domain tests were scored with a 4-point rubric (kappa = 0.81).

"Results from a paired sample t-test on pre–post tests showed that students obtained significantly higher domain test scores in the post-test (M = 15.79, standard deviation [SD] = 5.45) than the pretest (M = 13.50, SD = 5.26), t(33) = 2.171, p < 0.05, Cohen's d = 0.37."

Dina Soliman 2026 (2)¶

Dina Soliman, Xueqi Feng, Marlene Scardamalia, Carol Chan. (2026). Knowledge building–modeling: Grade 5 students creating models, using discourse analytics, and advancing systems thinking. Intern. J. Comput.-Support. Collab. Learn. https://doi.org/10.1007/s11412-026-09493-8

q2 · i? · design · r2

Methodological evidence on scoring: a '4-point rubric' anchored from limited to scientifically accurate and systemically reasoned, with two independent raters reaching kappa = 0.81.

"Responses to the domain tests were scored according to a 4-point rubric (1 = limited understanding; 4 = scientifically accurate and systemically reasoned). Two independent raters evaluated all responses with interrater reliability agreement (kappa = 0.81)."

Discussion¶

Learner Variables¶

• Prior Knowledge — outcome: instruction changes it

Related Claims¶

• Student model sophistication increased significantly across successive iterations during the KBM unit — related
• Qualitative analysis identified three mediating processes in KBM: discourse-driven idea improvement, analytics-supported meta-reflection, and model construction for system refinement — related
• Post-intervention domain knowledge correlated with final model scores and with higher-level systems thinking discourse (ST-3) — related
• Systems thinking discourse shifted away from lower-level moves toward higher-level interconnections and feedback moves across the KBM unit — related
• University students without qualitative-reasoning experience improved their Cerrado ecology domain knowledge after interacting with VISIGARP — related
• Students showed significant learning gains on domain knowledge (large effect) and concept maps (medium effect) after a 45-minute SRL-rich learning task — a broader claim this one bears on