Item-difficulty orderings diverge across cohorts, with the weakest correlation between academics and students (r = 0.188) and no stable shared developmental hierarchy¶
Claim · All claims
Evidence · 1 study (2 entries) · associationalr1–r2·q2·i1–i3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i1 The Spearman correlation between academics and students was r = 0.188 with a 95% CI spanning zero, indicating no stable shared developmental hierarchy. → Oliveira 2026
q2 i3 Academics and professional staff correlated moderately positively (r = 0.524, CI entirely above zero), suggesting a shared institutional logic. → Oliveira 2026 (2)
Evidence¶
Oliveira 2026¶
Oliveira, E., English, N., Ryan, T., Misiejuk, K., dal Ponte, C., López-Pernas, S., & Saqr, M. (2026). The GenAI Skill Bypass: Mapping Divergent Pathways of University Students and Staff AI Literacy. Preprint. https://arxiv.org/abs/2607.05411 doi:10.48550/arxiv.2607.05411
q2 · i1 · associational · r1
Spearman rank correlations on subgroup item-difficulty calibrations from the Guttman analysis (N=28 thresholds). The study reports "thecorrelationbetweenAcademicsandStudentsremainstheweakestofthethree(r= 0.188, 95% CI [-0.199, 0.524])", an interval spanning both negative and positive values.
"Third, thecorrelationbetweenAcademicsandStudentsremainstheweakestofthethree(r= 0.188, 95% CI [-0.199, 0.524])."
Oliveira 2026 (2)¶
Oliveira, E., English, N., Ryan, T., Misiejuk, K., dal Ponte, C., López-Pernas, S., & Saqr, M. (2026). The GenAI Skill Bypass: Mapping Divergent Pathways of University Students and Staff AI Literacy. Preprint. https://arxiv.org/abs/2607.05411. doi:10.48550/arxiv.2607.05411
q2 · i3 · associational · r2
Spearman rank correlation between academic and professional staff difficulty orderings (N=28 thresholds). The authors report "the moderate positive correlation between Academics and Professional Staff (r= 0.524, 95% CI[0.187, 0.750])", with the interval entirely above zero.
"the moderate positive correlation between Academics and Professional Staff (r= 0.524, 95% CI[0.187, 0.750])suggestsasharedinstitutionallogic."
Discussion¶
Related Claims¶
- Academic staff follow a traditional foundational-to-advanced GenAI skill trajectory, with creation items the least endorsed — related
- Students hit a metacognitive ceiling, endorsing high-level creation more frequently than the hardest metacognitive and evaluative thresholds in their ordering — related
- LLM-generated difficulty ratings show only moderate rank alignment with empirical student performance on basic arithmetic items — related
- Review-based AI overall difficulty correlates strongly with problem-level student pass rate across 79 parallel-class exam problems — related
- The AI review ruler's problem-level correlation weakens in an introductory course, where exam-level correlation is near zero and cohort composition dominates — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
cohort-difficulty-ordering-correlations-diverge |
title |
Item-difficulty orderings diverge across cohorts, with the weakest correlation between academics and students (r = 0.188) and no stable shared developmental hierarchy |
description |
Item-difficulty orderings diverge across cohorts, with the weakest correlation between academics and students (r = 0.188) and no stable shared developmental hierarchy |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
2 — oliveira-2026, oliveira-2026-2 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.