Explanation-seeking interaction depth is positively associated with task quality beyond baseline knowledge and volume¶

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

Subclaims¶

q2 i? In the parsimony regression controlling for Volume, Depth significantly predicted task quality (β = 6.27, p = .006), about 6.3 additional marks per SD on a 0–100 scale. → Vangelis Tsiligkiris 2026

Evidence¶

Vangelis Tsiligkiris 2026¶

Vangelis Tsiligkiris. (2026). What students ask matters: LLM interaction depth, task quality, and immediate recall in higher education. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-026-00617-6

q2 · i? · associational · r2

Hierarchical OLS regression with HC3-robust standard errors on 22 students' marked case analyses (blind-marked, ICC = 0.91). Depth was marginal before Volume was added (β = 5.29, p = .092) and "was significant" once Volume was controlled; Volume itself showed a marginal trend (β = 4.16, p = .051).

"In the parsimony model controlling for Volume, Depth was significant (β = 6.27, p = .006), indicating that a one standard deviation increase in explanation-seeking prompts was associated with approximately 6.3 additional marks on the 0–100 task scale."

Discussion¶

Related Claims¶

• Interaction depth is not associated with immediate recall or gains once baseline knowledge is controlled — related
• No reliable moderation of depth effects by baseline knowledge, though simple slopes suggest an exploratory pattern — related
• Scaffolding level provides no consistent additional predictive information about subsequent exam performance beyond prior achievement and observed dialogue behavior — related