A harm reduction approach to improving peer review by acknowledging its imperfections
Bibliographic record
Abstract
This candid perspective written by scholars from diverse disciplinary backgrounds is intended to advance conversations about the realities of peer review and its inherent limitations. Trust in a process or institution is built slowly and can be destroyed quickly. Trust in the peer review process for scholarly outputs (i.e., journal articles) is being eroded by high-profile scandals, exaggerated news stories, exposés, corrections, retractions, and anecdotes about poor practices. Diminished trust in the peer review process has real-world consequences and threatens the uptake of critical scientific advances. The literature on “crises of trust” tells us that rebuilding diminished trust takes time and requires frank admission and discussion of problems, creative thinking that addresses rather than dismisses criticisms, and planning and enacting short- and long-term reforms to address the root causes of problems. This article takes steps in this direction by presenting eight peer review reality checks and summarizing efforts to address their weaknesses using a harm reduction approach, though we recognize that reforms take time and some problems may never be fully rectified. While some forms of harm reduction will require structural and procedural changes, we emphasize the vital role that training editors, reviewers, and authors has in harm reduction. Additionally, consumers of science need training about how the peer review process works and how to critically evaluate research findings. No amount of self-policing, transparency, or reform to peer review will eliminate all bad actors, unscrupulous publishers, perverse incentives that reward cutting corners, intentional deception, or bias. However, the scientific community can act to minimize the harms from these activities, while simultaneously (re)building the peer review process. A peer review system is needed, even if it is imperfect.
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 | 0.349 | 0.414 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.009 | 0.005 |
| Science and technology studies | 0.019 | 0.090 |
| Scholarly communication | 0.040 | 0.036 |
| Open science | 0.009 | 0.030 |
| Research integrity | 0.017 | 0.035 |
| Insufficient payload (model declined to judge) | 0.005 | 0.004 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".