Critical thinking improvement differed by proficiency level: high-level students gained 18%, intermediate 12%, and low-level 8%, with asymmetric feedback adaptation¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? High-level students' critical thinking scores increased by 18%, low-level students by 8%, and intermediate-level students by 12%; feedback focused on structured scaffolds for low-level students and counter-questioning heuristics for high-level students. → Qi Zhu 2026

Evidence¶

Qi Zhu 2026¶

Qi Zhu, Xiaoming Zhai, Yan Zou, Chunlei Gao. (2026). Using AI-Generated Feedback to Improve Critical Thinking and Writing Proficiency. https://orcid.org/0000-0003-4519-1931 doi:10.1007/978-3-032-29788-4_37

q2 · i? · design · r2

Longitudinal comparative analysis of three proficiency cohorts stratified by pre-test scores, combining quantitative tracking and thematic analysis. The article reports the printed percentage gains and describes asymmetric adaptation: structured writing scaffolds for low-level students versus counter-questioning heuristic strategies for high-level students.

"high -level stu- dents' critical thinking scores increased by 18%, low -level students by 8%, and inter- mediate-level students by 12%."

Discussion¶

Related Claims¶

• A three-month WISE Agent intervention did not significantly increase sixth-graders' aggregate critical thinking disposition scores — related
• Some classroom discussion approaches improve text comprehension, but gains are not uniform across approaches and more student talk alone does not produce them — a broader claim this one bears on
• WISE Agent feedback significantly increased Truth-seeking and CT Self-confidence disposition scores — related
• BKT scaffolding significantly benefits low prior knowledge students, who outperformed low prior knowledge controls — related
• DRL scaffolding significantly benefits high prior knowledge students while BKT does not, and condition-by-prior-knowledge interactions are not significant — related
• Proportions of significant CT gains vary by instructional approach, but approach alone does not determine effectiveness — related
• The effect of mathematics test practice varies across prior-achievement levels, with significant effects at intermediate-to-high levels (levels 3 and 4) but not at the lowest or highest — related
• Prior vocabulary knowledge moderates vocabulary learning gains in an extensive reading program, with high-level learners gaining more than intermediate and low-level learners — related