Warning effect size shows no significant heterogeneity across topics and studies¶

Claim · All claims
Evidence · 1 study · causal r3 · q3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? The meta-analysis found no significant heterogeneity in warning effect size across the 32 topic/study cells. → Reed Orchinik 2026

Evidence¶

Reed Orchinik 2026¶

Reed Orchinik, David Rand. (2026). A light-touch AI literacy intervention helps protect against AI political persuasion. https://arxiv.org/abs/2609.16432 doi:10.48550/arxiv.2609.16432

q3 · i? · causal · r3

Heterogeneity test within the multilevel random-effects meta-analysis of warning effects across each randomized factor in each experiment (32 estimates). The printed test is Q(31) = 17.71, p= 0.97; no effect-size heterogeneity statistic is printed, so impact is null.

"We find no significant heterogeneity in warning effect size across topics/studies (Q(31) = 17.71,p= 0.97)"

Discussion¶

Related Claims¶

• AI literacy warnings reduce LLM political persuasion by roughly one-half (meta-analytic -48.1%) — related
• Effects of dropout programs on dropout odds are heterogeneous across studies — reports the opposite
• The overall effect of PBL in the tutor-background meta-analysis is modest (g = 0.27) with large, statistically significant heterogeneity — reports the opposite
• In Study 1, the general warning significantly reduced persuasive effects on housing-policy attitudes (b=−1.81, p= 0.001) — related
• Topic-wise gains were heterogeneous, with some topics showing no meaningful change — related