A Comparison of Machine-Graded (ChatGPT) and Human-Graded Essay Scores in Veterinary Admissions
Bibliographic record
Abstract
Admissions committees have historically emphasized cognitive measures, but a paradigm shift toward holistic reviews now places greater importance on non-cognitive skills. These holistic reviews may include personal statements, experiences, references, interviews, multiple mini-interviews, and situational judgment tests, often requiring substantial faculty resources. Leveraging advances in artificial intelligence, particularly in natural language processing, this study was conducted to assess the agreement of essay scores graded by both humans and machines (OpenAI's ChatGPT). Correlations were calculated among these scores and cognitive and non-cognitive measures in the admissions process. Human-derived scores from 778 applicants in 2021 and 552 in 2022 had item-specific inter-rater reliabilities ranging from 0.07 to 0.41, while machine-derived inter-replicate reliabilities ranged from 0.41 to 0.61. Pairwise correlations between human- and machine-derived essay scores and other admissions criteria revealed moderate correlations between the two scoring methods (0.41) and fair correlations between the essays and the multiple mini-interview (0.20 and 0.22 for human and machine scores, respectively). Despite having very low correlations, machine-graded scores exhibited slightly stronger correlations with cognitive measures (0.10 to 0.15) compared to human-graded scores (0.01 to 0.02). Importantly, machine scores demonstrated higher precision, approximately two to three times greater than human scores in both years. This study emphasizes the importance of careful item design, rubric development, and prompt formulation when using machine-based essay grading. It also underscores the importance of employing replicates and robust statistical analyses to ensure equitable applicant ranking when integrating machine grading into the admissions process.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.019 | 0.109 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".