Two editors from biomedical journals have found that journals with high-impact factors are more likely to publish retractions than those with lower impact factors.
Using the biomedical citation database PubMed, Infection and Immunity editor-in-chief Ferric C. Fang and mBio editor-in-chief Arturo Casadevall developed a “retraction index” by dividing the number of retractions by the number of total articles published by 17 scientific journals between the years 2001 and 2010. The journals selected had a wide selection of impact factors, ranging from 2.00 to 53.484, as calculated by Thomson Reuters Journal Citation Reports. Fang and Casadevall found a strong correlation between a journal’s retraction index and its impact factor.
Because high-impact journals focus on publishing influential research, these journals are more likely to publish questionable research that could potentially lead to paradigm shifts in their fields, according to Fang and Casadevall. While these journals have become targets for researchers who take short-cuts, withhold data, and distort images for personal gains such as job opportunities, grant money, and recognition, many retractions also occur because of innocent methodological errors.
“There’s a lot of emphasis on journals that retract articles because of scientific misconduct,” said Fang. “But a substantial proportion of articles are not retracted because of misconduct, but rather because of other aspects of the work that are called into question such as [an] incorrect reagent or a methodological problem–something that causes the authors to question their conclusions and makes them feel that it would potentially be misleading to the scientific community.”
But another major problem with these retractions is that they are not highly publicized.
Because a retracted article can remain in databases such as PubMed without any notice, such a paper could continue to acquire citations, driving up the impact factor of the journal that published it.
“It’s important to be transparent about the reasons for a retraction. If it was a bad reagent, what was that reagent? If there was a problem with the data, if there was a suspicion that the particular data were inappropriately manipulated—I think all of that information is in particular interest of the scientific community and should be made known,” said Fang.
Fang and Casadevall developed the retraction index after Infection and Immunity retracted six articles from a single laboratory earlier this year. These retracted articles included manipulated numbers, unreliable findings, and plagiarism.
“I think retractions, in almost all cases, are a tragedy even if there’s no misconduct,” said Fang. “Where a paper has to be retracted represents a huge amount of wasted effort and resources on the part of many people.”
The complete paper, “Retracted science and the retraction index,” was published August 8 in Infection and Immunity.