Make No Mistake

Can Scientific Journals Fix Their Mistakes?

It’s inevitable. Mistakes get published in scientific journals. But the real test is whether scientific journals can correct these mistakes. Writing in Nature today, David Allison and colleagues report that many journals may be failing this test. They identify and provide evidence of an urgent need for fixing the process for post publication review.

The issue is far from trivial. The authors identified 25 papers over a short period of time in obesity, nutrition, and energetics with factual errors sufficient to alter or invalidate the conclusions of those papers. They explain:

We showed that a small team of investigators with expertise in statistics and experimental design could find dozens of problematic papers while keeping abreast of the literature. Most were detected simply by reading the paper.

Fixing Publication ErrorsBut they found that correcting these problems is itself a considerable problem. They found onerous, ill-defined processes for raising concerns about such errors. In some cases, people who found the errors were asked to pay publication fees in order for their concerns to be reviewed. They found that examining the data supporting these papers was often difficult or impossible. Responses to concerns were slow. Retraction of invalid results was too often difficult or impossible.

The actions that would correct this problem are reasonably straightforward. Transparency is needed for data that support scientific publications. Responses to concerns about publication errors should be at least as swift as responses to submissions for publication. And finally, clear protocols should ensure that concerns about publication errors are handled efficiently and effectively.

The stakes are high. Scientific integrity demands a robust process for finding and correcting publication errors. The current barriers to this process must be removed.

Click here to read the commentary in Nature and here to read more from Retraction Watch.

Make No Mistake, photograph © theilr / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


February 4, 2016

5 Responses to “Can Scientific Journals Fix Their Mistakes?”

  1. February 04, 2016 at 6:21 am, Joe Gitchell said:

    This really highlights the importance of and how they need to keep growing and having greater influence.

    We all respond to carrots and sticks, and clearly journals could use a bit more of both!


    • February 04, 2016 at 7:36 am, Ted said:

      You’re right, Joe. When people review your work critically, they are telling you it’s important.

  2. February 04, 2016 at 10:01 am, Charles Baker said:

    Perhaps, the time has come for a journal’s’ impact factor to include its resolution/determinedness to correct/retract flawed articles.

    • February 04, 2016 at 3:39 pm, Ted said:

      Clearly, Charles, more attention is needed. Thanks!

  3. February 05, 2016 at 8:53 am, Allen Browne said:

    As the man said “Data and reason rarely overcome dogma and bias”. However this information is very valuable and points to holes in the”system” of peer reviewed papers.

    What is ? I’ll have to look that one up.

    Already learning this morning.