Corrections, Retractions & Expressions of Concern

Published papers do not stay frozen: mistakes are discovered, doubts arise and occasionally findings fall apart. Journals deal with this through public notices, not quiet edits. This guide covers research paper corrections, retractions and expressions of concern — the meaning of each notice, when it applies, how authors ask for a correction, and how readers verify a paper’s status before citing it.

The Reason Published Records Need Notices

After publication, an article becomes the version of record — the one others cite and build on. If it contains a mistake, the journal cannot quietly alter it: two readers citing “the same” paper would end up with different texts. Journals therefore issue a separate, permanent, linked notice. Knowing about these notices helps you correct your own papers and assess those you cite.

Five Notice Types

Certain publishers apply the word “correction” to errata and corrigenda alike.

Situations That Lead to a Correction

• A numerical mistake in results, a table or a figure that leaves the overall conclusions intact.
• A misspelt or missing author name, or an incorrect affiliation.
• A wrong or absent citation.
• A figure that is mislabelled, rotated or wrongly scaled.
• An omitted funding, ethics or conflict-of-interest statement.

Corrections do not cover rewriting for style, adding new results (which would be a new paper) or altering conclusions the authors no longer favour.

Choosing Between Correction, Concern and Retraction

What matters is how much of the paper is affected and whether a problem has been established. A portion of the paper: correction. Findings unreliable overall: retraction. Credible but unproven: an expression of concern while the facts are verified. The COPE retraction guidelines set out this approach for editors.

Grounds for Retraction

Honest Mistake Versus Misconduct

Retraction is not purely a punishment. Authors who find a serious error in their own work and request retraction are setting the record right — good practice that many editors and institutions respect. Notices should state plainly when a retraction results from an honest error. Saying so early usually earns more goodwill than waiting for someone else to raise it.

Expressions of Concern

An expression of concern appears when a serious, credible question exists but the investigation is still unfinished — for instance while an institution examines raw data. It can later be followed by a correction, a retraction, or its own withdrawal once the issue is settled.

Asking for a Correction

Move as soon as you notice an error. The corresponding author should write to the journal giving the paper ID, the precise location, the text as published, the corrected text and the reason, along with written consent from all co-authors. Never upload an altered version elsewhere and present it as the published article.

What Readers See

• An independent notice that is free to read, can be cited, and links to and from the article.
• Retractions carry a “Retracted” label on the article and in listings, and the original text normally stays readable.
• Refreshed metadata, letting databases and reference managers display the status.

Verify Before Citing

Prior to relying on an important paper, inspect its article page for notices. The Retraction Watch database, which Crossref acquired in 2023 and opened to everyone, logs retractions from many publishers, and several reference managers can flag retracted items. Should you need to cite a retracted paper, for instance to discuss it, say plainly that it was retracted. See how to write references.

Can a Correction Damage Your Reputation?

A prompt, honest correction generally reflects well on authors, showing concern for the record. Reputations suffer from concealing errors, or from misconduct. Retain raw data, analysis files and records so you can check and fix problems fast; see research paper authorship for how responsibility is shared among co-authors.

JMVRA’s Approach to Corrections and Retractions

The complete correction policy and retraction policy are available to read.

Share Your Research Through JMVRA

Your At-a-Glance Checklist

1. Act fastReport errors promptly
2. PreciseLocation and new text
3. AgreeAll co-authors
4. NoticeLinked, permanent
5. CheckStatus before citing
6. HonestSelf-correction respected

Practical Tips for Handling Errors After Publication

Discovering a mistake in a published paper is stressful, yet handling it openly is a sign of integrity. These suggestions help authors and readers respond calmly and constructively when the record needs repair.

• Describe the problem precisely. Before writing to the editor, identify the exact location of the error, what the correct information is and how you know. Precise reports are processed faster and are taken more seriously.
• Contact the editor promptly. Inform the journal as soon as you are confident of the issue instead of waiting for someone else to find it. Early disclosure shows good faith and limits how far the mistake spreads.
• Keep the original evidence. Preserve raw data, lab notebooks, scripts and earlier versions of files that support your account. If the journal asks questions, these records let you answer clearly and without guesswork.
• Inform your co-authors together. Tell all co-authors about the issue at the same time and agree on the wording of any request. A coordinated, joint message avoids contradictory statements reaching the editor.
• Look for notices before citing. When relying heavily on a paper, check its journal page for any attached notice. Building an argument on a withdrawn finding can undermine your own work and mislead your readers.

Report genuine errors quickly, support your claim with records, and always check a source's current status before building your own argument on it.

Questions Researchers Often Ask

In Summary

Corrections and retractions keep science self-correcting. Learn the five notice types, report your own errors fast with every co-author’s agreement, retain your data so problems can be checked, and confirm a paper’s status before relying on it. A transparent record benefits authors and readers alike. Treating a published error as a chance to improve the record, not as a threat, makes the whole process calmer.

References and Further Resources

• COPE retraction guidelines
• Crossref: Retraction Watch data
• JMVRA correction policy
• JMVRA retraction policy