Tuesday, April 05, 2011

Is science failing to self-police big claims?

What do you think happens when a splashy scientific claim is debunked by a rebuttal paper? Science self-corrects and the original claim fades away? Guess again.

An interesting new study says rebuttal papers have little impact on recent high-profile papers in fisheries science.

The biggest problem is that most rebuttal papers are ignored compared to the attention lavished on the original claims, based on a review of citation patterns. Rebuttals do matter when they get noticed; scientists citing rebuttals tend to be more criticial of the original claims.

Does this mean something is wrong with science? Since rebuttals "aim to highlight substantial flaws in published papers and act as the first line of defense after scientific research passes the review system," what should be done? The authors suggest:
"Our results point to an urgent need to change current publishing models to ensure that rebuttals are prominently linked to original articles."
This sounds like a big problem. Could it be that scientists are allowing fiction to parade as fact?

But there may be some self interest here. The authors of this paper are mostly on the "rebuttal" side of a divide that separates two competing research programs. The splashy papers come from some researchers who are challenging established views of fisheries science, and getting the expected rebuke.

So the identified "need" to change current publishing models looks like an attempt to get more attention to their own rebuttals.

I think people writing rebuttals should get attention focused on their rebuttals the old-fashioned way. Instead of seeking to change current publishing models, they should write a more compelling rebuttal.


Anonymous said...

The deeper question is this: Will anyone pay any attention to your critique of this paper?

Mark Powell said...

Looks like at least one person read it.