Retracted articles

I wonder if in some teaching sessions I ought to mention two more things, that relate to the literature the teaching hopefully helps students find.  

One thing is retracted articles.

The other thing is peer review.    

Actually in some sessions with first year students, we do talk about the peer review process, and I recommend they look to see if an information source has some sort of peer review process in place.     Maybe having a sort of process is better than not having one at all.  But I am coming across more about the shortcomings of the process.   

I occasionally review articles submitted to a library journal.   I was struck by the most recent one I reviewed, that the other reviewer spotted all sorts of things I did not.  Does that mean the process doesn't work?   Or that I am not a very good reviewer?    Or does it just show why you have more than one peer reviewer?   

Peer review and retractions go together.   The evidence from Retraction Watch is that the peer review process doesn't always work, as retracted articles have often gone through the process, which did not spot the problem.   Some retracted papers might have been written by a "paper mill", and others by generative AI.  Then, there are some that have used the more established methods of referencing irrelevant studies, using material like images without permission, manipulating images so they seem to show something else, making up results, or using suspect methods.  

I came across this example from Retraction Watch.   It would be a less contentious example of a retracted article than some, but it may not be the best teaching example for me, as it is a) in Dutch and b) made up (see below).    The story starts with this 2003 article in the Dutch language Nederlands Tijdschrift voor Geneeskunde, a long established general medical journal.

The article title is prefaced with "Clinical thinking and decision making in practice" (my translation).   The authors investigate an article that the same journal published in 1923, by H. van der Speck entitled "A case of urotopoeia" (Een geval van uroptoë - which a well known translation site put back into Dutch as "Een geval van utopie"!).   The title of the 2003 article ends (my translation) "the truth 80 years later".   A subscription is needed to read all the article, but Retraction Watch relates its finding that the 1923 article was a prank, submitted by a medical student, which the journal published.

The retraction, as related by Retraction Watch, does say that it's unlikely that the journal had a peer review system as we would know it, in 1923.    Although, presumably it did get past the editor.

I've read the suggestion that retracted articles are still cited after they are retracted, which is reason to talk about it.    

Has this one been cited?   In Scopus, only twice.    Once by the 2003 article that talks about it.   And once as a reprint of the original, published in the same 2003 issue of Nederlands Tijdschrift voor Geneeskunde.

So, this may not be a typical example of a retracted article.  But maybe it is a starting point to talk about the need to look out for retractions, and not use them in your work, and a way to talk about how to tell something in your results has been retracted, on a journal website (or more problematically, in a print journal), or through reference management software (Zotero, for example, works with Retraction Watch and flags things as retracted).