Why Most Published Research Findings Are False: Problems in the Analysis

  • Steven Goodman,
  • Sander Greenland
  • Published: April 24, 2007
  • DOI: 10.1371/journal.pmed.0040168

Reader Comments (2)

Post a new comment on this article

Clarification of an illogical conclusion

Posted by plosmedicine on 31 Mar 2009 at 00:08 GMT

Author: David Beebe
Position: Professor
Institution: Washington University
Submitted Date: April 25, 2007
Published Date: April 25, 2007
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

I was pleased to see this article appear in PLoS Medicine in response to the implausible article from Dr. Ioannidis that dealt with the perceived unreliability of medical research studies. In my view, Dr. Ioannidis’ paper did not pass that first test that I apply to any logical argument: reasonability.

A cursory read of Dr. Ioannidis’ paper revealed the he had chosen boundary conditions that seemed unreasonable, but that would assure that his argument would be validated. I am not a statistician or particularly adept at statistics and, therefore, did not write to challenge the conclusion of Dr. Ioannidis’ paper that most published research findings are false. However, as a practicing scientist, I knew that his conclusion was incorrect in most cases. It was a relief to see that Drs. Goodman and Greenland were willing to apply the expertise and to take the time to demonstrate the weakness of Dr. Ioannidis’ argument.

After reading Dr. Ioannidis’ article and the several laudatory responses to it, I was left with the question: "How many of the articles published in PLoS Medicine did the Editors think were false?” According to Dr. Ioannidis, the answer should have been “all or nearly all.” The Editors and most practicing scientists should have had a problem with that conclusion. However, in the face of such an apparent dilemma, the Editors afforded Dr. Ioannidis’ paper ‘top billing’ and permitted it to have a patently sensational title. In thinking of how such a logical conundrum could occur, I seriously considered that the Editors of PLoS Medicine were using Dr. Ioannidis’ paper as a ruse to see if the members of the scientific community could detect and respond to such an illogical conclusion. Although it has taken many months, it appears that they have.

No competing interests declared.