Last week Science retracted a 2011 paper written by a prominent social psychologist who has been accused of fabricating massive amount of data. Dr. Stapel is a prolific author and according to Retraction Watch, many more of his papers may be retracted as the case develops.
It seems that Dr Stapel is the protagonist of the biggest case of scientific misconduct in recent years, no wonder this is getting a lot of attention from the media. I am sure his tainted reputation is well-known by everybody working on the field of social psychology and researchers will think twice before citing any of Dr. Stapel’s paper on their own papers. But how about papers that were retracted but didn’t draw much attention? Retracting papers nowadays when most of the journals are available on line is not so difficult, but in 1999, Budd et al. analyzed retracted articles and found they received more than 2,000 post-retraction citations. Less than 8% of the citations acknowledged the retraction in any way. I believe this number would look much better. However, as Dr Sheldon Tobe, from the University of Toronto said, a study can be retracted, but “it can be hard to make its effects go away,”
More and more papers have been retracted in the recent years, some for honest mistakes and others for scientific misconduct. But the time between publication and retraction has also increased. The Wall Street Journal used the PubMed database and analyzed 742 articles in medicine and biology. Take the year of 2000, for example. Only 4 papers were retracted and it took in average 5.25 months between publication and retraction. In 2009, 179 papers were retracted and it took in average 32.62 month between publication and retraction.
If this exponential increase in paper retractions is a tendency, maybe in the future we will have a library of disgraced papers to consult before citing a paper.
Budd, John M., MaryEllen Sievert, Tom R. Schultz, and Caryn Scoville. 1999. “Effects of Article Retraction on Citation and Practice in Medicine.” Bulletin of the Medical Library Association 87(4):437–43.