A call to interpret disagreement components during classification assessment DATA.xlsx
Bibliographic record
Abstract
This foresight manuscript proposes several ideas concerning how to conduct insightful classification assessment. Authors should report disagreement components that relate to the research question, without anointing results as acceptable or good. This manuscript reviews the citations of the 2011 paper entitled ‘Death to Kappa: Birth of Quantity Disagreement and Allocation Disagreement for Accuracy Assessment’, which gave two recommendations: 1 do not use Kappa and 2 use disagreement components. We analyzed 200 articles that cited the Death to Kappa paper. A quarter of the articles followed both recommendations, another quarter followed only the first recommendation, another quarter followed only the second recommendation, and the last quarter followed neither recommendation. The attempt to replace Kappa with disagreement components has been partially effective, while Kappa continues to haunt several professions. We discuss misguided uses of Percent Correct and Kappa in Remote Sensing and Land Change Modeling. The concepts are general thus relate to additional fields. Authors frequently use arbitrary thresholds of metrics to claim that results are acceptable. However, the notion that results can be acceptable or not is inherently unscientific. Scientists must use a metric that addresses a clear research question in which the scientists have no vested interest in the results.
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 | 0.060 | 0.350 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.010 | 0.012 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.010 | 0.009 |
| Open science | 0.005 | 0.007 |
| Research integrity | 0.004 | 0.004 |
| Insufficient payload (model declined to judge) | 0.182 | 0.080 |
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".