Reference-table workshop: comparable quantities and missing context · RESEARCH-27596

Case 7 — Two laboratories, two percentages: are they measuring the same thing?

Before comparing laboratories, ask whether sample handling, method conditions and reporting definitions are available. Missing details should be visible in the table.

Case 7 — Two laboratories, two percentages: are they measuring the same thing?

Keep the original reported values beside any reformatted presentation. A conversion or interpretation needs a documented basis; it should not silently replace the source.

Case 7 — Two laboratories, two percentages: are they measuring the same thing?

An apparent difference can motivate a method question without proving that either laboratory is wrong. The appropriate next step is clarification from the reporting laboratory.

Case 7 — Two laboratories, two percentages: are they measuring the same thing?

Use a comparison-status column: directly comparable, comparable with stated qualifications, or not established. Explain the reason for each label.

Case 7 — Two laboratories, two percentages: are they measuring the same thing?

The summary should say that different quantities were initially mixed and identify how the revised table separates them. No clinical conclusion follows from this bookkeeping correction.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

The draft summary of a fictional reading exercise moves from an animal-model observation to a proposed human benefit. Which fields would keep the literature note within the evidence?

Exercise RESEARCH-27596/8: address this question within its stated limits.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

Put the species or model, design, comparator and measured endpoint before the interpretation. The reader should be able to see what was observed without relying on the headline.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

Was the analysis prespecified or exploratory, and how was uncertainty reported? Record what the paper states and mark unreported fields instead of assuming good practice.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

Separate the authors’ proposed mechanism from the measurements that support it. A plausible explanation is not automatically a demonstrated pathway.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

A human-benefit claim needs appropriate human evidence. This preclinical note should describe its model and translation limits rather than suggesting a treatment change.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

Check the primary paper, supplementary material and any publisher correction. A review article can help locate sources, but it is not a substitute for verifying the cited observation.

Case 8 — Reading a preclinical paper: what belongs beside the abstract?

The corrected summary can be useful without a clinical promise: model, methods, endpoint, uncertainty and remaining questions. Link each material statement to its source location.

Case 9 — A correction log that survives the next archive update

A archive table has a transcription error in a unit label. A later publisher notice also changes a figure. How can the correction history distinguish those two events?

Exercise RESEARCH-27596/9: address this question within its stated limits.

Case 9 — A correction log that survives the next archive update

Use separate entries for the internal transcription correction and the publisher notice. They have different sources and may affect different claims.

Case 9 — A correction log that survives the next archive update

Each entry should retain the date, old text, corrected text, reason and source reference. For a publisher notice, link the notice as well as the original paper.