The influence of rater training on inter-and intra-rater reliability when using the rat grimace scale
Bibliographic record
Abstract
Rodent grimace scales facilitate assessment of spontaneous pain and can identify a range of acute pain levels. Reported rater training in using these scales varies considerably and may contribute to observed variability in inter-rater reliability. This study evaluated the effect of training on inter-rater reliability with the Rat Grimace Scale (RGS). Two training sets, of 42 and 150 images, were prepared from several acute pain models. Four trainee raters progressed through 2 rounds of training, first scoring 42 images (S1) followed by 150 images (S2a). After each round, trainees reviewed the RGS and any problematic images with an experienced rater. The 150 images were then re-scored (S2b). Four years after training, all trainees re-scored the 150 images (S2c). Inter- and intra-rater reliability was evaluated using the intra-class correlation coefficient (ICC) and ICCs compared with a Feldt test. Inter-rater reliability increased from moderate (0.58 [95%CI: 0.43-0.72]) to very good (0.85 [0.81-0.88]) between S1 and S2b (p < 0.01) and also increased between S2a and S2b (p < 0.01). The action units with the highest and lowest ICCs at S2b were orbital tightening (0.84 [0.80-0.87]) and whiskers (0.63 [0.57-0.70]), respectively. In comparison to an experienced rater the ICCs for all trainees improved, ranging from 0.88 to 0.91 at S2b. Four years later, very good inter-rater reliability was retained (0.82 [0.76-0.84]) and intra-rater reliability was good or very good (0.78-0.87). Training improves inter-rater reliability between trainees, with an associated reduction in 95%CI. Additionally, training resulted in improved inter-rater reliability alongside an experienced rater. Performance was retained after several years. The beneficial effects of training potentially reduce data variability and improve experimental animal welfare.
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 | 0.184 | 0.287 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".