Carol Anne Meyer from CrossRef presented an update of her SSP 2011 presentation. After a Harvard researcher was accused of wrongdoing, his research paper was retracted by the publisher, but it remained on his website for several months. And it still can be retrieved on Google Scholar. It is also on PubMed (with a note about the retraction). There are multiple channels to access an article, which is a problem for users.
Users are not always clear which version of a document they are reading. Librarians do not have time to track changes after publication, so users might cite incorrect versions of articles. It is very tempting to go to the most accessible version of something. Researchers complain that Google does not retrieve the correct versions of articles.
How should we communicate article corrections? Sometimes journal editors issue an “Expression of Concern” or add a link to corrections on the web page for the article, such as this one:
But sometimes these links are difficult to find, at the bottom of the page for example. And the original PDF does not contain the corrections, so if the researcher downloads the original, the correction is lost. CrossRef has begun to address this system, putting a mark (CrossMark) on records that identify publisher-maintained content (see my posting linked above). A pilot test currently underway has shown that users do not know what the mark means, so CrossRef has added mouse-over information and a label explaining it. The mark can even be shown on PDF documents. The CrossMark service will be available in 2012. CrossRef is working with some search vendors to label articles in search results.
Columnist, Information Today and Conference Circuit Blog Editor