Instructor-confirmed grading errors occurred at roughly 0.02–0.04% of rubric line items across the term¶
Claim · All claims
Evidence · 1 study · designr2·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? Against roughly 6,000 rubric decisions per assignment, the 5–10 regrade requests per assignment, about half judged valid, correspond to an instructor-confirmed per-rubric-item error rate of approximately 0.02–0.04%. → Jesse A. Rodríguez 2026
Evidence¶
Jesse A. Rodríguez 2026¶
Jesse A. Rodríguez. (2026). LaTA: A Drop-In, FERPA-Compliant Local-LLM Autograder for Upper-Division STEM Coursework. arXiv preprint. https://arxiv.org/abs/2605.05410 doi:10.48550/arxiv.2605.05410
q2 · i? · design · r2
Regrade-request audit across the eight Winter 2026 assignments: the instructor classified each disputed rubric line item as valid or invalid. The quote carries the printed contested and confirmed per-item rates.
"Against this denominator, the 5–10 regrade requests received per assignment correspond to a contested per-rubric-item rate of 0.04–0.08% and an instructor-confirmed per-rubric-item error rate (after the ∼50% valid filter) of approximately 0.02–0.04%."
Discussion¶
Related Claims¶
- Students reported first-pass false-positive deductions at a mean of 2.94 per term, normalising to about 0.6% of rubric items — related
- The TA ground truth was produced under a one-rater-per-item design with co-grading, an exceptions log, instructor audit, and student review, with an estimated grading-error rate of about 0.1% in the student-inspected subsample — related
- An on-premises LLM autograder graded a 200-student course at 1–3 minutes wall-clock time per submission across a full term — related
- Students rated regular announcements/e-mail reminders and grading rubrics for all assignments as the most beneficial learner-to-instructor strategies — related
- Human Graders of Open-Ended Work Agree With Each Other Less Than Treating One Grade as Ground Truth Implies — related
- Instructor grading practices vary extremely, with some instructors awarding no honor grades and others awarding honor grades to over half their students — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
lata-instructor-confirmed-error-rate |
title |
Instructor-confirmed grading errors occurred at roughly 0.02–0.04% of rubric line items across the term |
description |
Instructor-confirmed grading errors occurred at roughly 0.02–0.04% of rubric line items across the term |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
generated |
by process:wiki-ingest on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — jesse-a-rodríguez-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.