In defense of preprints

At the end of July, I criticized an opinion piece that Tom Sheldon published in Nature, “Preprints could promote confusion and distortion“:

While the article casts preprints, preprint servers, and scientists who post their work to preprint servers as potential sources of misinformation, its arguments better support the case that science reporters should act more responsibly…What irks me is that I can’t find any hint in the article that Sheldon thinks journalists share the blame when they sensationalize bad science.

The latest issue of Nature includes three more critical responses to Sheldon’s nonsense.

James Fraser and Jessica Polka point out what Sheldon fails to mention, that journalists share responsibility for what they report (“Together scientists and journalists can spot poor preprints“):

Wherever they hear about a story, journalists are under the same obligation as scientists to critically review the work they intend to communicate to readers.

Jonathan Tennant, Laurent Gatto, and Corina Logan share this sentiment (“Preprints help journalism, not hinder it“):

A responsible journalist consults multiple independent sources to verify research findings. This critical evaluation is not contingent on the research having been peer reviewed.

Along the same lines, Sarvenaz Sarabipour (“Preprints are good for science and good for the public“) points out that

When scientists and journalists follow fundamental principles for reporting research results — such as ensuring that publications are rigorously sourced and fact-checked — preprints pose no greater risk to the public’s understanding of science than do peer-reviewed articles.

Dr. Sarabipour is also the lead author on a PeerJ preprint, “Maintaining confidence in the reporting of scientific outputs,” which makes a longer version of the same argument:

Our disagreement is rooted in recognizing that the tension between supporting preprints and good journalism is a false dichotomy. Limiting what or when preprints are released could suppress science communication without providing any clear benefit to the public. Instead, we argue that when basic journalism principles are followed, reporting on preprints should carry no greater risk to public understanding of science than reporting on peer-reviewed articles.

All of these arguments hammer home the same two basic points. First, Sheldon seems not to recognize that journalists are behaving irresponsibly when they report bad science through a failure of due diligence. Second, he provides no evidence whatsoever that preprints are more likely than peer reviewed articles to report incorrect results; in fact, the anecdotes he does provide make the opposite point.

Finally, I can’t help but mention, one more time, the utter cluelessness of Sheldon’s JAQing off:

Should all preprints be emblazoned with a warning aimed at journalists that work has not been peer reviewed?

You mean a warning like this?


Leave a Reply