JOURNALS CAN DEFINE PEER REVIEW HOWEVER THEY PLEASE, WITH NO MINIMUM STANDARDS NO ENFORCEMENT MECHANISMS.

The journal Science also published his work. In one instance, his paper was accepted after being examined by a single referee whose review consisted of a single paragraph. Reich writes: “Acceptance of a paper on the basis of only one review is a clear violation of Science’s published guidelines; the journal promises scientists, reporters, and other authorities that research has been reviewed by at least two experts.”

Conversely, Juan Miguel Campanario describes historical as well as contemporary referees who scoffed at work that later went on to win Nobel Prizes in chemistry, physics, and medicine.

If one gets what one pays for it’s worth noting that referees typically work for free. They lack the time and resources to perform anything other than a metaphorical sniff test. Nothing like an audit occurs. No one examines the raw data for she had personally encountered journal editors “who have almost boasted” about the fact that they themselves produce fake peer reviews rather than recruiting qualified referees.

Paraphrasing their remarks, she says such editors declare: “I never have a worry about finding reviewers because I just do it myself.”

Even when procedures are followed in good faith, the results don’t inspire confidence. A full 35 years ago, back in 1982, Douglas Peters and Stephen Ceci demonstrated peer review’s random, arbitrary nature by resubmitting papers to journals that had already published them.

Cosmetic changes (including the use of fictitious author and institution names) were made to 12 papers that had appeared in a dozen highly regarded psychology journals during the previous 18—32 months.

The duplication was noticed