Quality Costs: Gibberish, STEM Journals, and the Bottom Line

A 2009 study conducted by the National Humanities Alliance examined the flagship journals of eight scholarly organizations to weigh the cost of publishing humanities and social science research against the costs of publishing science, technology, engineering, and medical (STEM) research.  Mary Waltham, the consultant who authored the study, found that humanities and social science journals average $526 per page, a figure that doubles STEM journals’ average of $266.

When this study debuted five years ago, the muttering among my colleagues in “expensive” fields was that it only went to show that we do a far better job than the hastily cobbled-together, cheap productions sometimes issued by STEM fields.  Since then, we’ve been awaiting proof of our bias, and last month, we may have found it.  The journal Nature reported that the STEM-heavy publishers Springer and Institute of Electrical and Electronic Engineers (IEEE) had recently been forced to remove 120 nonsense papers from published journals.  A French computer scientist named Cyril Labbé had developed a program for detecting papers generated by software called SCIgen, much to the frustration of academic tricksters everywhere.  He has so far informed IEEE of 100 fakes and Springer of sixteen.  A least one of the papers was “authored” by scholars who had no idea their names were being used in such a way.

There are many stages in which an elementary standard of quality control could have eliminated nonsense papers from the pool—upon initial receipt, when the editors could review and desk-reject; at the peer-review stage, when a colleague could question authenticity; at the editing stage, when the copy editor could discover the gibberish.  It would appear that, in a minimum of 120 instances, at least two STEM publishers squandered repeated opportunities to provide their readership with genuine, useful research.

Advertisements