The AI review ruler's problem-level correlation weakens in an introductory course, where exam-level correlation is near zero and cohort composition dominates¶
Claim · All claims
Evidence · 1 study · associationalr2·q2·i2medium
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
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 i2 In the 106-problem CS101 sample, the problem-level correlation weakened to Spearman rho = -0.552, and the exam-level correlation across 16 examinations was near zero. → Hongfei Yan 2026
Evidence¶
Hongfei Yan 2026¶
Hongfei Yan, Jiangkai Xiong, Yiqing Li, and Chong Chen. (2026). From Evaluated Models to Evaluation Aids: A Multi-Evidence Study of LLM-Based Difficulty Calibration for Programming Examinations. https://arxiv.org/abs/2608.07523 doi:10.48550/arxiv.2608.07523
q2 · i2 · associational · r2
Boundary test of the same review pipeline on 106 problems from 16 past Introduction to Computing B final machine exams. Printed effect size: Spearman rho = -0.552 at the problem level. The article attributes the near-zero exam-level result to cohort composition rather than problem difficulty dominating exam-level outcomes.
"A 106-problem introductory-course (CS101) sample marks the boundary of the scale: the problem-level correlation weakened to rho = -0.552, and the exam-level correlation across 16 examinations was near zero, with co-"
Discussion¶
Related Claims¶
- The same review ruler supports longitudinal cross-semester problem tracking for one teacher's Data Structures and Algorithms B course — related
- The negative AI-difficulty–performance correlation is stable across exposure-discount values 0 to 0.40, and is strongest with no adjustment — related
- Review-based AI overall difficulty correlates strongly with problem-level student pass rate across 79 parallel-class exam problems — related
- Exam-level AI difficulty correlations with student performance are directionally consistent but weaker and only exploratory — related
- AI review difficulty also correlates with problem-level non-attempt rate, providing process evidence beyond pass rate — related
- Item-difficulty orderings diverge across cohorts, with the weakest correlation between academics and students (r = 0.188) and no stable shared developmental hierarchy — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
cs101-introductory-course-applicability-boundary |
title |
The AI review ruler's problem-level correlation weakens in an introductory course, where exam-level correlation is near zero and cohort composition dominates |
description |
The AI review ruler's problem-level correlation weakens in an introductory course, where exam-level correlation is near zero and cohort composition dominates |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — hongfei-yan-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.