Support This Website! Shop Here!

Monday, December 15, 2014

Peer Review Doesn't Work

I've seen people attempt to refute various scientific studies by saying, "They're not peer-reviewed, so they aren't any good."

I'm a little tired of that.
Click on each of the linked words below.

You will see peer review is not a gold  standard.
Click hereherehereherehere, or here.

In fact, it really isn't any kind of standard at all.
Click hereherehereherehere, here or here.

Take, for instance, the experience of John Bohannon, a biologist at Harvard. He submitted a faked paper on the effects of a chemical derived from lichen on cancer cells to 304 self-described “peer-reviewed” journals. The paper was completely fabricated: totally flawed design, analysis and interpretation of results. The researcher and the university were fictitious. 157 of the journals accepted it for publication. That's a failure rate of more than half.

In a classic 1998 study Fiona Godlee, editor of the prestigious British Medical Journal, sent an article containing eight deliberate mistakes in study design, analysis and interpretation to more than 200 of the BMJ’s regular reviewers. Not one picked out all the mistakes. On average, they reported fewer than two; some did not spot any. Peer reviewers also seem to get worse with experience.

Some people also claim that science is great because it depends on replication. If that's true, very little science is done anymore, because no one attempts to replicate studies anymore.

  • Replication is generally not publishable, so few people do it. 
  • Only people with an axe to grind pursue replications with vigour. 
  • More than half of 238 biomedical papers published in 84 journals failed to identify all the resources (such as chemical reagents) necessary to reproduce the results.
  • Five years ago about 60% of researchers said they would share their raw data if asked; now just 45% do.
  • Only 143 of 351 randomly selected papers published in the world’s 50 leading journals and covered by some data-sharing policy actually complied.
  • Some code used to analyse data or run models may be the result of years of work and thus precious intellectual property that gives its possessors an edge in future research.

So, the next time someone invokes "peer review", just shake your head.
They apparently don't read the journals.

Addendum:
From the comments in this link
Important scientific work has not uncommonly been initially rejected by peer-reviewed journals. As a 2001 article in Science observed, "Mention 'peer review' and almost every scientist will regale you with stories about referees submitting nasty comments, sitting on a manuscript forever, or rejecting a paper only to repeat the study and steal the glory."2 Indeed, an article in the journal Science Communication by Juan Miguel Campanario notes that top journals such as "Science and Nature have also sometimes rejected significant papers," and in fact "Nature has even rejected work that eventually earned the Nobel Prize."3 In an amusing letter titled "Not in our Nature," Campanario reminds the journal of four examples where it rejected significant papers: 
(1) In 1981, Nature rejected a paper by the British biochemist Robert H. Michell on signalling reaction by hormones. This paper has since been cited more than 1,800 times.
(2) In June 1937, Nature rejected Hans Krebs's letter describing the citric acid cycle. Krebs won the 953 Nobel prize in physiology or medicine for this discovery.
(3) Nature initially rejected a paper on work for which Harmut Michel won the 1988 Nobel prize for chemistry; it has been identified by the Institute of Scientific Information as a core document and widely cited.
(4) A paper by Michael J. Berridge, rejected in 1983 by Nature, ranks at number 275 in a list of the most-cited papers of all time. It has been cited more than 1,900 times.4
Second Addendum:
Psychologist "proves" ESP - a study in statistical insanity

Third Addendum:
 Neurogenesis in humans is crap

Fourth Addendum:
And most science that relies on statistical analysis cannot be reproduced. Another Sigh.

Fifth Addendum:
Retractions due to fraud are escalating at geometric rates

Sixth Addendum:
Is all of science a fraud?


Eighth Addendum:
And "reputable" journals are doing really lousy right now. 

Ninth Addendum:
Groups that detect scientific fraud are doing land-office business.

2 comments:

Tom Van Dyke said...

Turning over our society to the "experts" is fraught with danger, in no small part because

a) http://www.forbes.com/sites/gregsatell/2014/02/19/why-experts-always-seem-to-get-it-wrong/

b)
http://retractionwatch.com/

c)
Where are the conservative social psychologists?
http://www.psychologicalscience.org/index.php/news/were-only-human/74794.html

d)#MoreSocialThanScience [Twitter]

e)
http://www.patheos.com/blogs/warrenthrockmorton/2008/07/25/handicapping-the-apa-abortion-and-mental-health-task-force-report/

f)
http://www.heritage.org/research/reports/2012/10/the-regnerus-study-social-science-on-new-family-structures-met-with-intolerance

Etc. Further, the primacy of "science" outlaws even philosophy as not meeting "rational basis."

Read the bizarre story of how even Plato lost in court.

http://linguafranca.mirror.theinfo.org/9609/stand.html



Tom Van Dyke said...

See also

http://theweek.com/article/index/273736/how-academias-liberal-bias-is-killing-social-science