What Is the Difference Between a Correction, Erratum, Expression of Concern, and Retraction?

What Do These Publication Notices Actually Tell You?

You look up a paper and discover an unfamiliar notice beside it: Correction, Erratum, Expression of Concern, or perhaps Retracted. All four indicate that something happened after publication, but they do not communicate the same thing.

The distinction matters because researchers can make mistakes in both directions. You might reject a useful study simply because it has a correction, or continue treating seriously compromised findings as ordinary evidence because the original article remains accessible.

These terms are best understood as different mechanisms for maintaining and communicating the state of the scholarly record.

The Notices Represent Different Kinds and Degrees of Post-Publication Action

How the Four Publication Notices Differ

What Is a Correction?

A correction is a formal post-publication notice used to amend an error in a published work. The error may concern text, data, calculations, figures, author information, or other aspects of the publication.

The important point is that correction does not automatically mean invalidation. The International Committee of Medical Journal Editors notes that even pervasive inaccuracies caused by problems such as coding or calculation may be handled through correction when they do not change the direction or implications of the article's results, interpretations, and conclusions.

According to ICMJE recommendations, journals issuing corrections should identify the changes and cite the original publication, provide an updated version with the changes and dates documented, retain prior versions, and direct readers toward the more recent version. These recommendations are particularly relevant to medical journals, while practices in other disciplines and publishers can differ.

For you as a reader, the essential question is therefore not simply, "Has this paper been corrected?" It is whether the correction affects the evidence you want to use.

What Is an Erratum?

Erratum is another term used for a published correction. Historically and across publishers, terminology has not been perfectly standardized. Some journals or publishers have distinguished an erratum from a corrigendum according to who introduced the error, while others use broader correction terminology.

That variation makes it risky to infer the seriousness or origin of an error from the word erratum alone. An erratum may concern something minor, but the appropriate interpretation comes from reading the notice itself.

Bibliographic systems may also preserve terminology that differs from what a publisher currently uses. PubMed, for example, maintains links between original articles and associated errata and other publication notices. This is one reason researchers should focus on the substance of the linked notice rather than expecting every journal and database to use identical labels.

What Is an Expression of Concern?

An expression of concern is fundamentally different from a correction. It does not simply say, "Here is an identified error, and here is the fix." Instead, it warns readers that potentially significant questions about a publication remain unresolved.

COPE's retraction guidance recognizes expressions of concern as an interim action that editors may use in appropriate circumstances while an investigation or disciplinary process is unresolved. ICMJE similarly notes that an editor may publish an expression of concern while allegations or concerns about the integrity of published work are being investigated.

An expression of concern is therefore neither a declaration that the article has been proven invalid nor reassurance that the article is sound. It communicates uncertainty serious enough that readers should know about it.

If a paper you intend to rely on carries such a notice, the next question is how the unresolved concern should affect your use of the paper.

What Is a Retraction?

A retraction is a formal mechanism for correcting the literature when a publication should no longer be relied upon in the ordinary way because of serious problems.

COPE advises editors to consider retraction when there is clear evidence that findings are unreliable because of major error or fabrication or falsification, as well as in circumstances involving plagiarism, redundant publication, unauthorized material or data, serious legal issues, unethical research, compromised or manipulated peer review, or certain undisclosed competing interests that materially affected interpretation or editorial recommendations.

This means that retraction is not synonymous with research misconduct. A paper can be retracted because of an honest but consequential error. Conversely, retraction can result from misconduct or other serious breaches. The retraction notice should explain why the action occurred.

COPE recommends that retraction notices clearly identify the affected publication, state who is retracting it and why, be freely available, and remain linked to the retracted article. The purpose is to correct the scholarly record, not simply to make the original publication disappear.

These Notices Are Not a Simple Ladder of Severity

It is tempting to imagine a neat sequence: erratum, then correction, then expression of concern, then retraction. Scholarly publishing is not that tidy.

A correction and an erratum may be alternative terminology rather than different levels of seriousness. An expression of concern can precede a retraction, a correction, another resolution, or no immediate further action, depending on what an investigation establishes. A retraction can arise from honest error rather than misconduct.

The Reason Matters as Much as the Label

Two papers carrying the same type of notice may require very different responses. A correction to an author's institutional affiliation has little bearing on whether an experimental result is valid. A correction to the primary outcome calculation may be much more consequential.

Likewise, two retractions may have very different histories. One might follow discovery of an honest analytical error that invalidates the conclusions. Another might involve fabricated data. Both are retractions, but understanding why they occurred is essential if you are discussing their methodological, historical, or research-integrity implications.

This is why researchers should check important papers for post-publication notices and then read those notices rather than treating the presence or absence of a label as a complete quality assessment.

Four Similar-Looking Papers, Four Different Situations

The notices do not merely assign different labels to "bad papers." They tell you different things about what is known, what has changed, and what remains uncertain.

Common Misinterpretations of Publication Notices

Respond to the Substance of the Notice

When you encounter any post-publication notice, separate two questions: What editorial action has occurred? and what does that action mean for the evidence I intend to use?

When You Encounter a Post-Publication Notice

Questions About Corrections, Errata, Concerns, and Retractions

The Label Tells You What Happened, Not Everything It Means

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.