Average homework H-scores correlate negatively with proctored exam performance, with the correlation strengthening from Spring 2024 to Spring 2026¶
Claim · All claims
Evidence · 1 study · associationalr2·q2
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 i? The per-semester correlation between average H-score and exam performance was near-zero in 2020–2023 and strengthened to a strongly negative value by Spring 2026. → David Racovan 2026
Evidence¶
David Racovan 2026¶
David Racovan, Ajay Rawat, Christopher K. May, Jeffrey A. Turkstra. (2026). Argus: Academic Integrity in the Era of Generative AI. https://arxiv.org/abs/2609.36073 doi:10.48550/arxiv.2609.36073
q2 · i? · associational · r2
Longitudinal observational analysis of six offerings comparing averaged homework H-scores against closed-note, proctored exam percentages per student; the text reports a negative relationship emerging in Spring 2024 and strengthening through "Spring 2026 (𝑟=− 0.537)". Spring 2020 was limited because a project replaced exams beyond Midterm 1.
"Beginning in Spring 2024, a negative relationship begins to emerge (𝑟=− 0.244), and strengthens in Spring 2025 (𝑟=− 0.450) and Spring 2026 (𝑟=− 0.537)."
Discussion¶
Related Claims¶
- Students received significantly higher homework scores during GenAI-evaluation periods, but this advantage did not transfer to downstream exams — the general claim this finding bears on, with the studies behind it
- Average H-scores rose from near-zero in pre-LLM offerings to 11.9 in Spring 2026, with maximum scores reaching 81 — related
- Random measurement noise in pretest scores produces a negative contribution to the correlation between normalized gain and pretest score — related
- High-ability students' learning behaves as a dominant α-process even at low pretest scores, while average-ability students show more γ-process behavior — related
- Students who withdrew after Midterm 1 had higher H-scores than course completers, with the gap widening by Spring 2026 — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
h-score-negative-exam-correlation-strengthening |
title |
Average homework H-scores correlate negatively with proctored exam performance, with the correlation strengthening from Spring 2024 to Spring 2026 |
description |
Average homework H-scores correlate negatively with proctored exam performance, with the correlation strengthening from Spring 2024 to Spring 2026 |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
generated |
by process:wiki-ingest on 2026-10-10 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — david-racovan-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.