Cracking the code: residents’ interpretations of written assessment comments
Bibliographic record
Abstract
CONTEXT: Interest is growing in the use of qualitative data for assessment. Written comments on residents' in-training evaluation reports (ITERs) can be reliably rank-ordered by faculty attendings, who are adept at interpreting these narratives. However, if residents do not interpret assessment comments in the same way, a valuable educational opportunity may be lost. OBJECTIVES: Our purpose was to explore residents' interpretations of written assessment comments using mixed methods. METHODS: Twelve internal medicine (IM) postgraduate year 2 (PGY2) residents were asked to rank-order a set of anonymised PGY1 residents (n = 48) from a previous year in IM based solely on their ITER comments. Each PGY1 was ranked by four PGY2s; generalisability theory was used to assess inter-rater reliability. The PGY2s were then interviewed separately about their rank-ordering process, how they made sense of the comments and how they viewed ITERs in general. Interviews were analysed using constructivist grounded theory. RESULTS: Across four PGY2 residents, the G coefficient was 0.84; for a single resident it was 0.56. Resident rankings correlated extremely well with faculty member rankings (r = 0.90). Residents were equally adept at reading between the lines to construct meaning from the comments and used language cues in ways similarly reported in faculty attendings. Participants discussed the difficulties of interpreting vague language and provided perspectives on why they thought it occurs (time, discomfort, memorability and the permanency of written records). They emphasised the importance of face-to-face discussions, the relative value of comments over scores, staff-dependent variability of assessment and the perceived purpose and value of ITERs. They saw particular value in opportunities to review an aggregated set of comments. CONCLUSIONS: Residents understood the 'hidden code' in assessment language and their ability to rank-order residents based on comments matched that of faculty. Residents seemed to accept staff-dependent variability as a reality. These findings add to the growing evidence that supports the use of narrative comments and subjectivity in assessment.
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 | 0.081 | 0.307 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.005 | 0.012 |
| Scholarly communication | 0.006 | 0.006 |
| Open science | 0.002 | 0.008 |
| Research integrity | 0.002 | 0.004 |
| 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".