In a paper published this week in Infection and Immunity (IAI) by Ferric Fang and Arturo Casadevall, “Retracted Science and the Retraction Index.”, the authors have calculated what they are calling the “retraction index” for various scientific journals and plotted it against the impact factors for those same journals:
(Image linked from a post on Retraction Watch. h/t DrugMonkey)
The retraction index for a journal is the number of retractions from 2001 to 2010, times one thousand, divided by the number of published articles. Although the authors are careful to not claim a causal relationship between impact factor and retraction index, they are clearly implying that the higher retraction indexes of high impact journals represents something bad about those journals relative to other journals.
Here is an alternative hypothesis. The reason high impact factor journals have more retractions than low impact factor journals is that there is much less chance of getting away with shitte in a publication in a high impact factor journal: fucketonnes more people read the shitte and fucketonnes more people try to build on the shitte. The low impact factor literature is riddled with bogus shitte that never gets retracted because no one reads itte, no one tries to build upon itte, and no one gives a shitte.
9 comments
4 pings
Skip to comment form ↓
Freerefill
August 11, 2011 at 12:12 pm (UTC -7) Link to this comment
Case and point: Remember that Journal of Cosmology paper that claimed to find fossilized bacteria in a meteorite fragment? That didn’t get retracted, as far as I am aware, despite the OVERWHELMING amount of peer-reviewed reasons why it was wrong.
Besides, retractions are part of science. Analyzing, correcting, and growing is fundamental. To analyze which journal has more retractions is a bogus study; if they claim to be doing real science, they should ALL have retractions.
DrugMonkey
August 11, 2011 at 12:53 pm (UTC -7) Link to this comment
Of course you have no basis for this other than knee jerk defense. Please explain the threshold number of people reading a paper to detect fraud. And per usual, you ignore the simple fact that contingencies *matter* in driving human behavior.
http://scienceblogs.com/drugmonkey/2009/10/responsible_conduct_of_researc.php
DJ
August 11, 2011 at 2:27 pm (UTC -7) Link to this comment
Hey asshole, the “alternative hypothesis” that you proudly propose is right in the fucken manuscript:
“Alternatively, publications in high impact journals have increased visibility and may accordingly attract greater scrutiny that results in the discovery of problems eventually leading to retraction. It is possible that each of these explanations contributes to the correlation between retraction index and impact factor. Whatever the explanation, the phenomenon appears deserving of further study.”
Isis the Scientist
August 11, 2011 at 4:37 pm (UTC -7) Link to this comment
Hey asshole, the “alternative hypothesis” that you proudly propose is right in the fucken manuscript:
HA HA HA HA HAHA!!!!! I like the commenters over here. They seem less likely to take your bullshit at face value.
Nentuaby
August 11, 2011 at 5:15 pm (UTC -7) Link to this comment
It seems like it would even be the reverse- the contractions would be a causal factor in the high impact? After all, fuck(e)ing up is an everyday occurrence pretty much all ’round. Prompt and conscientious retraction of the fuckups is a part of keeping high quality standards, which surely have at least *some* impact on Impact.
Reginald Selkirk
August 11, 2011 at 7:34 pm (UTC -7) Link to this comment
Personally I’m waiting for the ultimate irony: will Fang & Casadevall (2011) ever have to be retracted?
Janne
August 11, 2011 at 8:19 pm (UTC -7) Link to this comment
I wonder if the really high impact factor glamour journals (sorry, “really impacte fuckutore” on this blogg(e) I guesse) can really be treated in the same (“samme”?) way as the rest(e). They tende to pubblisshe a differente typee* of paper than the more, well, detailed, nitty-gritty oriented papers of everyday science oriented journals. For instance, they tend to run a lot of rushed-to-print “we’ve discovered a new planet|geological formation|genera|particle” type papers that tend to get falsified with later observations.
* “Typee — cover that embarrassing lack of substantive ideas in your prose! Get your typee today!”
DrugMonkey
August 11, 2011 at 8:28 pm (UTC -7) Link to this comment
Freerefill and Janne, do be careful not to conflate being wrong with data faking. Two different issues entirely. I happen to think that being wrong is not grounds for retraction, merely for a correction. I think retraction should be for faking, falsification or fraud.
Janne
August 11, 2011 at 8:38 pm (UTC -7) Link to this comment
DrugMonkey: I agree fully. An honest but wrong paper is, together with a later refutation, _very_ valuable; it tells others “we tried this and got fooled by this sneaky factor. In reality it didn’t work. Don’t do the same mistake”.
There is a largish gray area where you can question if they were honestly wrong, sloppy, or willfully ignored things they shouldn’t in order to get the publication. But the balance should be on keeping the paper around, together with its refutation.
Not how it works today, unfortunately. Since journals or article databases often don’t do an explicit mention in the online version to later information there’s a real risk people who find it will believe the results still stand.
This week in science - Online Political Blog
August 13, 2011 at 6:06 pm (UTC -7) Link to this comment
[...] professor at a private medical school who blogs about politics, sports, food, science, booze and academia, and not necessarily in that [...]
อาหารเสริมบํารุงผิว.com
December 7, 2011 at 1:20 am (UTC -7) Link to this comment
อาหารเสริมบํารุงผิว.com…
[...]Correlation And Causation | Comradde PhysioProffe[...]…
The Greek Yogurt Diet
December 7, 2011 at 6:26 am (UTC -7) Link to this comment
The Greek Yogurt Diet…
[...]Correlation And Causation | Comradde PhysioProffe[...]…
flowers delivery usa
December 7, 2011 at 9:00 am (UTC -7) Link to this comment
flowers delivery usa…
[...]Correlation And Causation | Comradde PhysioProffe[...]…