End User Tools for Evaluating Scholarly Content

The existence of multiple versions of scholarly content (from author websites, institutional repositories, government archives, subject-specific digital libraries, aggregator collections and publisher websites) make it difficult for users to locate the most recent version of a document or to ascertain if the document has had any updates or even been retracted. This session describes tools for end users to evaluate the content they come across to make sure they are citing the most authoritative version of the content available. The reader will learn about the CrossMark version of record service and the importance of educating users about how to locate current information. What Happens When Scholarly Content Changes? Marc Hauser was a well-known primate researcher at Harvard University. He was on the faculty in the Psychology Department, ran the Cognitive Evolution Laboratory, and was a popular teacher and a leader in his field. Not unusually, the lab had its own web site, and publications of affiliate faculty and researchers were listed and copies were hosted on this site. One of the papers on the site, a 2002 article co-authored by Hauser, became the focus of a year-long Harvard University ethics investigation. The results of the investigation were unfortunate for Hauser. The paper, originally published in the Elsevier journal Cognition, was retracted, the case was made public in the Boston Globe, he was suspended, and he ultimately left Harvard. Published scholarly research is supposed to be selfcorrecting. But what happens when articles that have been corrected, updated or even retracted, as in the case of this paper, are still available in their original form? I first came across the case of Marc Hauser in December 2010. At that time, the retracted paper was still available on the Cognitive Evolution Labs Web site as a PDF. The article was formatted for publication in Cognition, and was clearly labeled “Article in Press.” There was no mention of a retraction or any investigation. In preparation for an article I was writing and several subsequent presentations, I checked the lab’s site in January, March, June, and November of 2011. It wasn’t until November that the PDF was removed from the lab’s site. In fact, the entire site had disappeared by that time. At no time was there any indication that the paper had been retracted. If you follow the CrossRef DOI link to the paper, you find that the publisher has clearly marked it “Retracted” by adding the word to the title. So the researcher who is using the publisher version will pretty clearly be able to learn about the retraction. If, on the other hand, a researcher starts at Google, and we heard at another session at this conference a study of librarians, faculty and students that reminds us that that is exactly what academics do, and he or she types in a search for the article “Rule Learning by Cotton-top Tamarins” the first result is the PDF of this article, on another site (Citeseer), which does not indicate the article has been retracted. The second result is to PubMed, which does have a way to notify researchers of retractions, but not of other types of corrections. Of course Google results vary. When I retried this search in preparing this article for the proceedings version of this talk, the first result was in a group from Google Scholar, and had a link to the properly identified retracted article at Cognition, but one of the other 13 results for the same citation was a stillavailable PDF of the paper on the co-author’s departmental web site at New York University. v In today’s world scholars have a broad array of sources for the research they rely on, from author home pages, to institutional repositories, web aggregators, government repositories, multiple ebook formats and readers, and federated search engines. This array is convenient, but it also presents challenges for notifying readers when something has happened to a document after peer review and publication. The Concerns of Librarians About a year ago, CrossRef conducted several focus groups with librarians. We wanted to know whether they had concerns about different versions of arti-