The within-teacher demand–AI-use association differs sharply across tasks, with only a modest positive planning gradient¶

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

Subclaims¶

q2 i? The three task-specific slopes differ significantly (omnibus Wald test F(2, 48) = 52.53, p < .001). → Holster 2026 q2 i1 The planning slope is modestly positive (OR = 1.084), with predicted planning use rising 3.18 percentage points across a two-point relative-demand span. → Holster 2026 (2)

Evidence¶

Holster 2026¶

Holster, J. (2026). Where Artificial Intelligence Enters Teacher Work. https://www.oecd.org/en/data/datasets/talis-2024-database.html doi:10.35542/osf.io/a39rp_v1

q2 · i? · associational · r2

Omnibus design-based Wald test of the task × task-relative-demand interaction in the primary pooled allocation model; the article reports "F(2, 48) = 52.53, p < .001" with 48 denominator degrees of freedom from the replicate design.

"The modest overall allocation association masked substantial differences across tasks, F(2, 48) = 52.53, p < .001."

Holster 2026 (2)¶

Holster, J. (2026). Where Artificial Intelligence Enters Teacher Work. https://www.oecd.org/en/data/datasets/talis-2024-database.html. doi:10.35542/osf.io/a39rp_v1

q2 · i1 · associational · r2

Task-specific slope for lesson planning from the same pooled model; the article reports "OR = 1.084, 95% CI [1.009, 1.164]" and a standardized probability difference of +3.18 points (95% CI [0.36, 6.01]) from Table 2.

"The primary system-average slope was modestly positive for lesson planning (OR = 1.084, 95% CI [1.009, 1.164], p = .028; Holm-adjusted p = .030)"

Discussion¶

Related Claims¶

• Assessment/marking shows an inverse demand–AI-use association in the primary system-average analysis, with predicted use falling 3.47 percentage points across a two-point relative-demand contrast — related
• Overall demand–AI-use alignment varies modestly across education systems, but task-specific association profiles show clear residual heterogeneity — related
• The prespecified task–demand mapping produces the strongest within-teacher alignment of all six possible one-to-one assignments, supporting task-specific rather than arbitrary demand alignment — related
• Teachers experiencing higher average demand across lesson planning, assessment/marking, and special-education adaptation report higher odds of AI adoption — related
• Among AI-using teachers, tasks experienced as relatively more demanding are more likely to be AI-supported overall — related
• Special-education support/adaptation shows the strongest demand–AI-use alignment, with a 16.32-percentage-point increase in predicted use across a two-point relative-demand contrast — a narrower finding that bears on this claim
• AI-related professional learning does not clearly moderate the within-teacher demand–AI allocation association — related