Sunday
Jan152012
by 
Bishop Hill
Bishop Hill Nature: British science needs integrity overhaul
Nature magazine has picked up on the BMJ survey of research misconduct in the UK - I posted about this a couple of days ago. The article carries the bold title British science needs `integrity overhaul'.
British scientists are fundamentally failing to deal with research misconduct, which is widespread in the country, leading experts have warned.
At a conference in London yesterday, participants were united in calling for more action on the issue.
“There is a recognition that we have a problem,” said Fiona Godlee, editor-in-chief of the British Medical Journal (BMJ) and one of the driving forces behind the meeting.
I think it would probably help if journals like Nature stopped trying to cover up for the miscreants.
Reader Comments (11)
Someone calling himself "Bishop Hill" posted the following comment at Nature:
"Nature has been at the forefront of efforts to cover up the misdemeanours of scientists involved in Climategate. This article therefore rings very hollow."
Very apt.
I agree that the article by Nature rings very hollow. It is the height of hypocrisy.
Lord Justice Brian Leveson, Nov 2011:
“The press provides an essential check on all aspects of public life. That is why any failure within the media affects all of us. At the heart of this Inquiry, therefore, may be one simple question: who guards the guardians?”
My allegation of fraud against Phil Jones is the sole explicit substantiated allegation of research fraud against any climate researcher in the UK, AFAIK. Nature reported on my allegation on 15 February 2010, in an article entitled “‘Climategate’ scientist speaks out”. The article was seriously inaccurate.
I filed a formal complaint against Nature with the Press Complaints Commission (PCC). Nature subsequently agreed to append the following statement to the online version of article:
That is nowhere near enough; so I pressed the PCC for more. At the time, I was so naive, I thought that the PCC actually did what they are supposed to do. In fact, the PCC ruled that Nature had not breached the Code of Conduct, and need not do anything—so of course nothing was done.
The whole reason that GLP and GMP was introduced into the fields of human and veterinary pharmaceutical development, was to deal with the abuses that occurred under the 'self-regulating' integrity of the medical and scientific 'professionals' involved. Naturally it was assumed that no scientist or medic would allow profit or company, or personal gain to over-ride 'ethics' or 'standards'. Thalidomide and other episodes proved this to be false. The problem with climatology is that it is currently unregulated, or, at least, only pal-reviewed. The trouble is, how do we sort this regulation problem at a national, regional or world-wide level?
Once a journal has been politicized, its ethics have been unconsciously compromised beyond recovery, since political adherents tend to judge themselves by their intentions, not by their actions nor by the results. With complicity of the PCC, the editors can remain in denial long past the point where their actions have resulted in death or serious losses for a great number of people. This will end very badly unless the PCC does what it is supposed to, and very soon.
I beleve that fraud and a failure to correct innocent mistakes could be quickly stamped out if a few brave journals were willing to adapt the following simple rules:
(1) All data, computer programs, methods and meta data must be archived in a publicly accessable, AUDITED database, BEFORE any articles are accepted for submission for peer review.
(2) Before publishing, any necessary corrections stemming from peer review must also be archived in the same manner.
(3) All prior papers referenced in such articles must also comply with these regulations. (This may require a grandfather rule, but with a strict cut off date).
Once a few solid journals accepted these rules, then the cowboy publications would be forced to reform or would quickly fade away.
This process of reform would be given a trernendous boost, if all professional members of scientific societies wrote to the editors of their journals requesting that such regulations be adopted.
Introducing integrity by edict or by wishing it into existance is just too hard.
What is required are rules that are seen to be enforced by being audited by independent chartered accountants.
Such rules must provide for automatic access to data and methods, so those with a doubting or sceptical turn of mind can turn the work upside down, looking for faults that may or may not exist.
Then scientists will rush to be seen to be honest, even handed, fair and careful, because it will pay them so to do.
They may even be willing to enlist the assistance of professional statisticans in the original design of their experiments and studies and in the verification of their results.
Jan 15, 2012 at 7:42 PM | Salopian
"The trouble is, how do we sort this regulation problem at a national, regional or world-wide level?"
Begin by recognizing that physicians are not scientists but care givers who do not regulate themselves in accordance with scientific method or its ethics. The BMJ serves physicians and serves science only secondarily. Keep in mind that there are enormous opportunities for profit in suffering and the vast majority of physicians are interested primarily in the profitable suffering.
The implementation of scientific method is trivially easy. Require everyone to share all their raw data, methods, and computer code for each article they publish. That would end 99% of all fraud in science, including physician researchers but not the other physicians. This step has always been obvious. What has been lacking is the political will.
Physicians who are not researchers, at least 99% of them, care diddly for scientific method. Surgeons invent the new technique, apply it, and if it is successful they market it widely. The science follows later if at all. At this time, there are no clear guidelines on how to regulate physicians or BMJ, excepting physician researchers and their research articles.
The 'internationally renowned' journal has a few interesting quotes. They appear to be wanting to point out that British science is the worst for this.
"British scientists are fundamentally failing to deal with research misconduct, which is widespread in the country, leading experts have warned."
"Research-integrity issues in the United Kingdom have long been fretted over. Last year the House of Commons science and technology select committee said that they found “the general oversight of research integrity in the UK to be unsatisfactory”"
Nature now has a follow-up editorial further debating the research fraud topic, blaming UK libel laws for suppressing whistleblowers. Important comments in thread also.
http://www.nature.com/nature/journal/v481/n7381/full/481237b.html?WT.ec_id=NATURE-20120119
Hat-tipped from Hilary's place
http://hro001.wordpress.com/2012/01/18/nature-says-face-up-to-fraud/