Word count correlates with suggested final grade, with the ordering Gemini Pro > humans > ChatGPT > Gemini among statistically significant correlations (human Spearman's rho = 0.603)¶
Claim · All claims
Evidence · 1 study · designr2·q3·i3large
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.
q3 i3 Among statistically significant correlations between word count and final grade, the ordering was Gemini Pro > humans > ChatGPT > Gemini, with the human correlation at Spearman's rho = 0.603. → Brevik A 2026
Evidence¶
Brevik A 2026¶
Brevik A, Jerpseth H and Lafontan SR (2026). Large language models as grading assistants in public health education: a method-comparison study of essay-style exam assessment. Front. Educ. 11:1904452. https://doi.org/10.3389/feduc.2026.1904452
q3 · i3 · design · r2
Spearman correlation analysis between submission word count and final grade (Table 4). The discussion reports the printed effect size "Spearman's rho = 0.603" for humans (p < 0.001), with Gemini Pro at 0.642, ChatGPT at 0.501, and Gemini at 0.495; Kimi's 0.311 was not significant (p = 0.083).
"Whether the slightly higher than expected human grade and word count correlation (Spearman's rho = 0.603) indicates that this metric can serve as a benchmark metric in attempts at LLM-human grading alignment remains to be confirmed in larger studies."
Discussion¶
Related Claims¶
- Claude showed the highest alignment with human BREQ responses, and interview-containing prompts aligned better than baseline prompts — related
- LLM grading patterns cluster into six groups aligned with vendor families, with highest correlations within same-vendor model families — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
wordcount-grade-correlation-llm-human |
title |
Word count correlates with suggested final grade, with the ordering Gemini Pro > humans > ChatGPT > Gemini among statistically significant correlations (human Spearman's rho = 0.603) |
description |
Word count correlates with suggested final grade, with the ordering Gemini Pro > humans > ChatGPT > Gemini among statistically significant correlations (human Spearman's rho = 0.603) |
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 — brevik-a-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.