Erratum or retraction?
Where a problem has been reported, and an action is required, you have a number of options. The most appropriate course of action depends on what the error is, why it occurred and (most importantly) whether it affects the conclusions of the article. The points below should help you decide which course of action is most appropriate.
The first thing to do is to consider what problem has been reported. A problem that might seem important to the author or other person reporting it may actually be quite trivial and not warrant any correction or action. If you decide this is the case, and that no post-publication correction is required, you should inform the person who reported the problem, giving your reasons.
When considering what action to take, note that BioMed Central adheres to a permanency policy. This states that corrections to, or retractions of, published articles will be made by publishing an erratum or retraction notice and without altering the original article in any way other than to add a prominent link to the original article.
- Erratum: An erratum appears as a new article, citing the original published article and correcting an error in it. Note that whilst BioMed Central uses the terms “Erratum” other publishers use “Correction” or “Corrigendum”.
- Retraction: Retractions are published to set the record straight. Where there are severe errors in an article that invalidate its conclusions, the article may need to be retracted.
And remember that if a problem is trivial, such as a small typo that does not change the meaning of the results/conclusions, there is always the option to do nothing.
Sometimes authors want to update their articles. In most cases it would be best for the author to submit the update either as a new manuscript, or as a letter to the editor (or other similar article type).
If you receive a (relatively) minor correction from the author very soon after publication, do you think is it OK to simply correct the online article and not publish an erratum? Absolutely not!
Whilst it can be tempting to do this, it must not be done for several reasons:
- BioMed Central deposits all articles within several safe open access archives. These ensure that the published article will be archived safely, and always be available (even if BioMed Central is sold, closes, or has all its servers stolen). Therefore any correction would mean replacing the deposited articles – not only logistically difficult, but also not permitted in some systems.
- At the moment of publication alerts are sent and readers will find the article through online searches: therefore no matter how fast a change is made, it is always possible (or even likely) that someone will have read the article, and possibly downloaded it – therefore integrity of the article will be invalidated if it changes on the BioMed Central site.
- It may also be important to alert readers to the correction so that any citations or subsequent referral display exactly what was changed and when.