I deeply appreciated the articles "Naughty Scientists" (C&EN, June 27, 2005, page 50) and "Research Misconduct" (C&EN, Aug. 29, 2005, page 24), which set out a relevant question in science. They were followed by an article on the precise question of peer review in medicine (C&EN, Sept. 19, 2005, page 32). Intense competition between researchers and managerial demands surely play important roles in misconduct, but I want to go further into this topic and consider the reasons for and the ways to correct these deficiencies.
Consider the process of publication of a paper. There are three groups of people who intervene along this process: the author, the editor, and the peer referee. When people do their work properly, it would be impossible to end up with a bad paper published with errors in it.
But there is also an intense competition among editors who need many papers to develop or maintain their journals. Some editors have a list of various specialists from various countries on their editorial or advisory boards only for show. The problem with this may appear when the editors do not have a submitted paper appropriately refereed. Thus, after a few tedious trials and unpleasant refusals, authors who want to publish a lot of papers after a short study know which journals will accept their papers and publish them quickly without any refereeing consideration.
Experimentalists are often able to carry out nice experiments, but their problem appears when they have to put their data through a mathematical treatment. Equations that are tedious to use are in the literature along with numerical methods that are built by the theoreticians who want to express facts well. Sometimes, instead of using these equations or models, the experimentalists yield to the temptation to select models that are easier to handle, even if they are not appropriately applicable.
Moreover, what is the answer to this question: Do the data still belong to the authors when they are published, and could the authors refuse to give additional information on the way they obtained their results?
Playing the role of a referee in various journals, I am wondering about this dilemma: Am I fair to require that a paper be reconsidered and revised by the authors, for the reason that they use a wrong equation, or at least not the best one, when I know that another journal will accept that paper without difficulty?
J. M. Vergnaud