Science 2.0: Thomson Reuters Pledges More Impact Factor Transparency
Impact factor, the relative authority of a journal based on article quality, is maligned by people who can't get into top-tier publications but though it was originally created to help libraries know which journals are worth paying for, articles in high impact factor journals are now considered a barometer of quality for government funding agencies, who can't be tasked with analyzing the merits and editorial policies of 25,000 different publications.
The system isn't perfect but greater transparency is one of the hallmarks of Science 2.0. I wrote in the Wall Street Journal about the plight of PNAS, the Proceedings of the National Academy of Sciences, and their papers that were clearly not peer reviewed in their quest to push out 10,000 articles a year, but Public Library of Science publishes 3X that in just one imprint, PLOS ONE, and some of the articles they publish are clearly just a compilation of anecdotes where a friendly editor checked off four boxes and the company ran the credit card. Those are exceptions but imagine the problems at smaller groups desperate for money or when unethical people try to game the system.
Thomson Reuters has released its Journal Citation Reports for 2014, using 2013 citation data, and also says it will increase transparency by providing citation-based metrics for articles, not just journals, along with some insight into their secret sauce for impact factor. It should now be possible for any subscriber to duplicate the work of Thomson Reuters - replicability was important in many ways, I noted in the WSJ - and it will be possible to normalize the counts across disciplines that may cite more heavily.
They also added 39 journals to the "Title Suppressions" list - they will not get an impact factor, because of anomalous citation patterns, basically cheating such as stacking and self-citing. Law Library Journal, for example, had 90% self-citing. And Richard Van Noorden at Nature notes that one researcher has been busted for citation stacking two years in a row. University of Alabama computer scientist Yang Xiao is not doing any favors for the Crimson Tide by being the poster child for unethical journal behavior.
Thomson Reuters says it is making modifications due to customer requests - complainers on the Internet who don't even subscribe are not making a big difference - and that what people do with impact factors may be bad, but that isn't the fault of the methodology, any more than it is the fault of a spoon for making someone fat.
Transparency promised for vilified Impact Factor by Richard Van Noorden, Nature
Image: D. Leonard Corgan Library, Kings College, Wilkes-Barre