Wednesday, April 18, 2012

NYT: The rise in The Rise in Research Retractions .

"The higher a journal's impact factor, the two editors found, the higher its retraction rate."

To be fair, retraction rates have to do with work that can be sloppy, or even work that was done with all due diligence but for reasons that were unknowable at the time turned out to be wrong.  But many of the retractions are outright fraud.  The main highlight of the article is a 10 fold increase in retractions over a decade when publishing had only gone up 44%.

The author of the article highlights what happens when unscrupulous people seek attention AND also believe they won't be caught with regard to the paper publishing industry.  A bigger issue might be the proposal writing industry that draws in the $billions and isn't nearly so well monitored.  Then there are the non-falsifiable sciences where proving someone was wrong is impossible, thus, making fraud a theoretical impossibility.


Delirious said...

The teachers at our high school have some kind of program that alerts them if the student has used text from a previously written assignment. My neighbor is a teacher there, and she said the program works pretty good. Why don't the universities use this? Or do they?

Looney said...

I have heard that universities do use plagiarism checkers. From my perspective, this only catches the least sophisticated cheaters. Plagiarism also doesn't introduce new error into the sea of information that we call science.
A far bigger problem is a researcher who will run some experiments and then either change the result or filter them in order to obtain a support or refute a particular theory.