In an OLS regression, explicit finish lines in teacher prompts were associated with smaller DOK gaps, while explain-reasoning epistemic framing was associated with larger gaps¶

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

Subclaims¶

q2 i? Prompts specifying explicit task endpoints (β=−0.22, p<.001) were associated with smaller DOK gaps. → Alex Liu 2026 q2 i? Prompts requesting students to explain reasoning were associated with larger DOK gaps (β=+0.20, p<.001), which the authors attribute to residual confounding with target ambition. → Alex Liu 2026 (2)

Evidence¶

Alex Liu 2026¶

Alex Liu, Min Sun, Lief Esbenshade, Victor Tian, Zachary Zhang, and Kevin He. (2026). Teacher-Authored Prompts for Configuring StudentAI Dialogue: K12 Classroom Implementation. https://arxiv.org/abs/2604.16738 doi:10.48550/arxiv.2604.16738

q2 · i? · associational · r2

OLS regression of DOK gap on prompt characteristics (N = 1,362, robust SEs clustered by discussion; R² = 0.21). The article reports the finish-line coefficient β=−0.22, p<.001, an unstandardized coefficient with no printed effect size.

"Prompts specifying explicit task endpoints (β=−0.22,p<.001) were associated with smaller DOK gaps."

Alex Liu 2026 (2)¶

Alex Liu, Min Sun, Lief Esbenshade, Victor Tian, Zachary Zhang, and Kevin He. (2026). Teacher-Authored Prompts for Configuring StudentAI Dialogue: K12 Classroom Implementation. https://arxiv.org/abs/2604.16738

q2 · i? · associational · r2

Same OLS model (Table 5). The article reports the explain-reasoning coefficient β=+0.20, p<.001, and cautions the positive coefficient may reflect residual association after controlling for target DOK rather than a negative effect of epistemic framing.

"Epistemic framingshowed counterintuitive effects, as prompts requesting students to "explain reasoning" were associated withlargerDOK gaps (β= +0.20,p<.001)."

Discussion¶

Related Claims¶

• Instructionally supported (prompted) journal writing outperforms unsupported journal writing on learning outcomes (Hedges's g = 0.78) — the general claim this finding bears on, with the studies behind it
• A design-enactment gap in cognitive demand: 37.7% of conversations under-reached the teacher-targeted DOK level, and under-targeting increased with target ambition — related
• Teachers authoring student-AI dialogue prompts predominantly write highly specific tasks targeting higher-order thinking (92% at DOK 2–3) — related
• LLMs show a directional asymmetry in educational control: they reliably increase cognitive demand but struggle to lower it — related