ChatGPT feedback quality varied by essay quality: accuracy declined as essay quality rose, prioritization was best for average essays, and supportive tone was lowest for low-quality essays¶
Claim · All claims
Evidence · 1 study · causalr2·q3
Currency · newest study 2023 · not yet checked against its sources · not reviewed by a person · recheck by 2026: 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 i? Across low-, average-, and high-quality essays, AI feedback differed significantly on accuracy (p < .001), prioritization of essential features (p = .03), and supportive tone (p = .001). → Steiss 2023
Evidence¶
Steiss 2023¶
Steiss, Jacob; Tate, Tamara; Graham, Steve; Cruz, Jazmin; Hebert, Michael; Wang, Jiali; Moon, Youngsun; Tseng, Waverly; Warschauer, Mark. (2023). Comparing the Quality of Human and ChatGPT Feedback on Students' Writing. https://eric.ed.gov/?q=Comparing+the+Quality+of+Human+and+ChatGPT+Feedback+on+Students+Writing
q3 · i? · causal · r2
ANOVA across three essay-quality groups found AI accuracy ratings of 4.42, 4.05, and 3.31 for low, average, and high essays; "the accuracy of AI feedback declined as the quality of the underlying essay rose." Prioritization means were 2.98, 3.22, 2.69; supportive tone 3.56, 4.1, 4.16.
"Specifically, the accuracy of AI feedback declined as the quality of the underlying essay rose, with ratings of 4.42, 4.05, and 3.31 for low-, average-, and high-quality essays, respectively."
Discussion¶
Related Claims¶
- Human evaluators' prioritization of essential features was significantly better for low-quality essays than for average- and high-scoring essays — related
- Essay quality correlates positively with the affective feature of ChatGPT feedback and negatively with the affective feature of peer feedback — related
- Trained human evaluators provided higher-quality formative feedback than ChatGPT on four of five components, while ChatGPT outperformed humans on criteria-based feedback — related
- Differences between human and ChatGPT feedback quality were small, with means within one point and partial eta-squared ranging from .01 to .12 — related
- No significant relationship exists between essay quality and the overall quality of feedback generated by peers and ChatGPT — reports the opposite
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
ai-feedback-quality-varies-by-essay-quality |
title |
ChatGPT feedback quality varied by essay quality: accuracy declined as essay quality rose, prioritization was best for average essays, and supportive tone was lowest for low-quality essays |
description |
ChatGPT feedback quality varied by essay quality: accuracy declined as essay quality rose, prioritization was best for average essays, and supportive tone was lowest for low-quality essays |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2023 |
recheck_by |
2026 |
generated |
by process:wiki-ingest on 2026-10-15 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — steiss-2023 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.