12 July 2012

Retraction tracking

I’ve always thought that the purpose of retracting a paper was to expunge it from the scientific literature. To pretend it was never published at all. To me, that always implied that a retracted paper should not even be citable, let alone cited.

Reading a post on the Retraction Watch blog, I’m realized that my view was too simple. In response to me, Ivan Oransky wrote:

There’s nothing wrong with citing a retracted paper, and that’s why good retraction practice is to leave it in the record but mark it “retracted.” The problem is when retracted papers are cited as if they were never retracted(.)

Journals should develop a standard for clearly indicating when a cited paper has been retracted. Journals already have exacting formatting for citations, so it should be easy enough for them to add an example like this.

“If the paper you are citing has been retracted, this should be indicated by placing an ‘r’ after the year in the text citation, and ‘(Retracted.)’ following the title in the full citation.

“For example:

“(Jones, 1980r)

“Jones J. 1980. Adverse artefacts of arachodoic acid. (Retracted.) Chinese Journal of Inexplicable Results 5:12-15. Retraction notice: 6:4”

This would give people a clear indication that citing a retracted paper is allowable (hopefully as a warning to others rather than to back a position). It might also help to emphasize that authors, reviewers, and editors should check for the retracted status of papers they cite. And it would increase transparency all around.

I’m totally doing this when I take over as the Editor-in-Chief of the Chinese Journal of Inexplicable Results.

External links

Despite refutation, Science arsenic life paper deserves retraction, scientist argues

No comments: