AI-peer integrated feedback yields significant gains on all four writing dimensions, with the group × dimension interaction showing differential improvement favoring task achievement and lexical resource over coherence and cohesion¶

Claim · All claims
Evidence · 2 studies · 2 causal r1–r2 · q3 · i3 large
Currency · newest study 2026 · checked against its sources 2026-10-19 (automated, 2 of 2 entries) · not reviewed by a person

Subclaims¶

q3 i3 The group × writing dimension interaction was significant (partial η2 = 0.024), and the mixed-design ANCOVA also showed a significant main effect of group (partial η2 = 0.26), indicating the intervention's effectiveness varied across the four writing criteria. → Liu W 2026 q3 i3 Simple effects favored the experimental group on all four dimensions, with effect sizes descending from task achievement (d = 1.41) to grammatical range and accuracy (d = 1.16). → Liu W 2026 (2)

Evidence¶

Liu W 2026¶

Liu W (2026) AI-peer integrated feedback in second language writing classes: exploring students' engagement and writing performance. Front. Psychol. 17:1871951. https://doi.org/10.3389/fpsyg.2026.1871951

q3 · i3 · causal · r2

A 2 (group) × 4 (writing dimension) mixed-design ANCOVA with composite pretest writing score as covariate; the omnibus interaction tested whether intervention effectiveness varied across the four IELTS-based writing criteria. The main effect of group was also significant (partial η2 = 0.26).

"Most importantly, group × writing dimension interaction reached statistical significance (F(2.85, 342.00) = 2.93, p = 0.036, partial η2 = 0.024, 95% CI [0.001, 0.07]), justifying follow-up simple-effect analyses"

Liu W 2026 (2)¶

Liu W (2026) AI-peer integrated feedback in second language writing classes: exploring students' engagement and writing performance. Front. Psychol. 17:1871951. https://doi.org/10.3389/fpsyg.2026.1871951

q3 · i3 · causal · r1

Bonferroni-adjusted follow-up simple-effect analyses on the mixed-design ANCOVA. Task achievement and lexical resource gains significantly exceeded coherence and cohesion gains (p = 0.023 and p = 0.031); coherence and cohesion showed "the smallest yet still significant gains".

"All four dimensions showed significant between-group differences favoring the experimental group, with effect sizes descending as follows: task achievement (d = 1.41, 95% CI [1.04, 1.77]), lexical resource (d = 1.28, 95% CI [0.92, 1.63])"

Discussion¶

Related Claims¶

• AI-peer integrated feedback produces significantly higher posttest behavioral engagement than conventional peer feedback in EFL writing — related
• In one quasi-experiment in EFL writing classes, students given AI-peer integrated feedback reported higher posttest cognitive engagement than students given conventional peer feedback — related
• Gains from automated scoring concentrated in linguistic accuracy and logical coherence but were absent in information fidelity and delivery fluency — related
• Students reported recognizing writing improvements after revising in response to AI feedback, with observed rubric-score gains on resubmission — related
• AI-assisted peer feedback enhanced lexical diversity only in the first writing task, with the group advantage disappearing by the second task — related
• AI-peer integrated feedback produces significantly higher posttest affective engagement than conventional peer feedback in EFL writing — related
• Performance during the AI-supported task consistently predicted performance on the subsequent AI-removal task across all writing dimensions — related
• Generative AI-assisted feedback is associated with a statistically significant pretest–posttest improvement in violin students' intonation performance over a four-week period — related
• Neither feedback mode produced differential improvements in syntactic complexity; a general time effect appeared only in the first task — related