Donate to Remove ads

Got a credit card? use our Credit Card & Finance Calculators

Thanks to johnstevens77,Bhoddhisatva,scotia,Anonymous,Cornytiv34, for Donating to support the site

The Thorny Issue of Fraudulent Research

Scientific discovery and discussion
modellingman
Lemon Slice
Posts: 614
Joined: November 4th, 2016, 3:46 pm
Has thanked: 595 times
Been thanked: 364 times

The Thorny Issue of Fraudulent Research

#39023

Postby modellingman » March 16th, 2017, 9:38 am

An interesting Guardian comment piece on some of the realities of current life in academia concerning fraudulent research, including manipulation of results, self-plagiarism (regurgitating one's own previous research findings as new research) and other sins. The author's organisation is not indicated in the article but he appears to have the same name as a lecturer in business ethics at Cardiff University.

https://www.theguardian.com/higher-educ ... -solutions


Apart from the "publish or perish" pressure that the writer mentions in the article, it seems to me that there are two other problems here:

- the sponsorship of research puts pressure on researchers to find the "right answers" - ie those that are aligned with the interests and biases of the sponsoring organisation. Whilst this is probably less of a problem in the "harder" sciences, I suspect it is more of a problem in "softer" sciences and particularly in social science research. The discussion concerning the PACE trial in the comments section following the article containers a pointer in this direction. I have also heard of this sort of pressure from a senior academic acquaintance and economist who has undertaken research for government related to skills and their impact on the UK's economic productivity.

- the cosy and self-regulating nature of the academic journals through which research is often disseminated is, I suspect, also an issue. Although, it is a good few years since I was actively involved, I have been a referee for a well-regarded journal run by a learned society. The society does extremely well financially from its journals and over my career (I'm now retired) I've seen its principal journal move from quarterly to monthly publications and its range of journals expand from one to five. Unusually, perhaps, for a referee of an academic journal, I was an industrial practitioner rather than a university academic but my observation was that journal editors didn't usually have an excessive amount of choice (or even any real choice) when appointing referees for specific submissions and, given the relatively small numbers of people involved in the specialised areas of the field where I did referee there was a very good chance that I would know of, if not actually know, the authors whose work I was being asked to judge. Whilst the society would no doubt vigorously defend the quality of its journals one can't help but wonder whether some compromises are arising, perhaps unintentionally, to protect the publishing cash-cow.

The article's writer suggests a kind of internal affairs unit but makes absolutely no suggestion about how this might be arranged. Perhaps at one stage the research councils might have been an appropriate home for this kind of quality control but I suspect this hasn't been the case for a very long time.

tjh290633
Lemon Half
Posts: 8209
Joined: November 4th, 2016, 11:20 am
Has thanked: 913 times
Been thanked: 4097 times

Re: The Thorny Issue of Fraudulent Research

#39047

Postby tjh290633 » March 16th, 2017, 10:56 am

I think that the key to this is efficient refereeing, preferably by someone in an unconnected organisation with experience in the field.

Like you I have done some refereeing, usually in a field in which I did some research back in the 1960s, but which I left for Industry soon after. The main offence that I found was plagiarism, using a graph from another's work without acknowlegement. It was a subject with which I had kept up, and so it was easy to spot that sort of event.

The use of "Cook's Constant", or selective choice of results by others, was another thing that occurred. Unfortunately it seems to frequently get by the referees in certain controversial fields.

Self-plagiarism I do not understand. Citing previous articles by one's self is fine, but most journals will only publish original research unless it is a review article, or similar. You would expect a referee familiar with the field to spot that.

TJH

modellingman
Lemon Slice
Posts: 614
Joined: November 4th, 2016, 3:46 pm
Has thanked: 595 times
Been thanked: 364 times

Re: The Thorny Issue of Fraudulent Research

#39232

Postby modellingman » March 17th, 2017, 9:32 am

Whilst I agree that an effective refereeing system is the answer as far as journals go, my own experience suggests that from a practical perspective the task of obtaining truly knowledgeable and independent referees may not be that easy. There are financial pressures on journal editors to attract and get content out and career pressures on academic researchers to get work published.

However, journals are not the only means of publishing. Because it is so long since I had an involvement, I have no idea how (or even if) the system of refereeing fits into the new world of open and self-publishing.

I also suspect that the bigger problems lie with commissioned research: there are a number of university groups around the country, particularly in the fields of social research and economics, that rely on commissioned research work for an appreciable slice of their income. Such work is often commissioned by government (and often through a competitive tendering exercise) but sometimes by other organisations such as trade associations. One of the reasons that such groups gain these commissions is that they lend an air of academic respectability to the eventual findings.

The safeguards of refereeing are generally not present in this type of commissioned work and commissioners can exert influence on outcomes a) through their choice of research group b) by influencing control during the research itself: for example, by closing down emergent questions c) by leaning on researchers at the report writing stage to emphasise or de-emphasise particular aspects and d) through contractual arrangements which retain control over whether and how the work is published. Whilst this may not result in fraudulent research in the sense of fiddled results, it is less than honest and provides a less than sound basis for any policy work (development, effectiveness monitoring, etc) that the commissioner may use it for.

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: The Thorny Issue of Fraudulent Research

#39261

Postby XFool » March 17th, 2017, 11:40 am

modellingman wrote:An interesting Guardian comment piece on some of the realities of current life in academia concerning fraudulent research, including manipulation of results, self-plagiarism (regurgitating one's own previous research findings as new research) and other sins. The author's organisation is not indicated in the article but he appears to have the same name as a lecturer in business ethics at Cardiff University.

https://www.theguardian.com/higher-educ ... -solutions

In case anyone hasn't seen it, in that Guardian article there was a reference to Robert Millikan. An associated link led to this interesting piece in the New York Review of Books from 2004:

http://www.nybooks.com/articles/2004/11 ... n-science/

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: The Thorny Issue of Fraudulent Research

#41596

Postby XFool » March 27th, 2017, 9:53 am

More on the subject of 'Fake research' in the UK in a BBC article: http://www.bbc.co.uk/news/science-environment-39357819

Kantwebefriends
Lemon Slice
Posts: 356
Joined: November 5th, 2016, 4:02 pm
Has thanked: 26 times
Been thanked: 102 times

Re: The Thorny Issue of Fraudulent Research

#41619

Postby Kantwebefriends » March 27th, 2017, 11:10 am

modellingman wrote:An interesting Guardian comment piece on some of the realities of current life in academia concerning fraudulent research


It's a bit rich for the Guardian to carry that, given its devotion to the fakery of "Climate Science".

Anther area of conspicuous fakery is the promotion of the "lipid hypothesis" about the causes of cardiovascular diseases.

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: The Thorny Issue of Fraudulent Research

#41746

Postby XFool » March 27th, 2017, 9:51 pm

Kantwebefriends wrote:
modellingman wrote:An interesting Guardian comment piece on some of the realities of current life in academia concerning fraudulent research

It's a bit rich for the Guardian to carry that, given its devotion to the fakery of "Climate Science".

Oh Gawd! I thought we'd lost all that nonsense...

UncleEbenezer
The full Lemon
Posts: 10692
Joined: November 4th, 2016, 8:17 pm
Has thanked: 1459 times
Been thanked: 2965 times

Re: The Thorny Issue of Fraudulent Research

#45808

Postby UncleEbenezer » April 13th, 2017, 4:51 pm

tjh290633 wrote:Self-plagiarism I do not understand. Citing previous articles by one's self is fine, but most journals will only publish original research unless it is a review article, or similar. You would expect a referee familiar with the field to spot that.

TJH

Could that be talking about a researcher's incremental update? Present something once, develop it further, present it again. It's clear in the author's mind that the second paper adds significantly to what the first one said, but perhaps doesn't stand on its own without repeating material from the first. A critic might take a different view.

modellingman wrote:I also suspect that the bigger problems lie with commissioned research:

You mean as in https://www.youtube.com/watch?v=G0ZZJXw4MTA ?
That is probably overrepresented among what is widely reported, precisely because its whole purpose is to support an Agenda.

Digging a bit further, there's an issue of what is flavour of the times and most likely to attract research funding. This is where the climate denialists have a point: not all the research carried out under that banner is of the best quality. And of course, the journos just love to jump on all that is most dodgy.

XFool wrote:Oh Gawd! I thought we'd lost all that nonsense...

I only just clicked this thread (following the NASA one), as I had previously been put off this board by that gibbering, and a thread title that looked suspiciously like a cover for more of it.

Historic comparisons in research volume and quality are bound to find changes due to the sheer volume of "publish or perish", and the growth of a long tail of publications whose name sounds serious, but who solicit content by spamming people like me for articles with a remit so wide as to be bound to encompass something I might have published when I was in that game a generation or so ago. I think the most promising answer to that is automated Peer Review using AI/Big Data, which (if you strip out real or imagined personalisation) is the Holy Grail of Google's ranking algorithms.

tjh290633
Lemon Half
Posts: 8209
Joined: November 4th, 2016, 11:20 am
Has thanked: 913 times
Been thanked: 4097 times

Re: The Thorny Issue of Fraudulent Research

#46473

Postby tjh290633 » April 17th, 2017, 7:00 pm

UncleEbenezer wrote:
tjh290633 wrote:Self-plagiarism I do not understand. Citing previous articles by one's self is fine, but most journals will only publish original research unless it is a review article, or similar. You would expect a referee familiar with the field to spot that.

TJH

Could that be talking about a researcher's incremental update? Present something once, develop it further, present it again. It's clear in the author's mind that the second paper adds significantly to what the first one said, but perhaps doesn't stand on its own without repeating material from the first. A critic might take a different view.


Most people will refer to their previous work, rather than repeat it all verbatim. I've seen plenty of updates to previously published work, sometimes in the form of a letter to the Editor.

If it were a radical change to a procedure previously used, then I can see why there might be a comparison of the two, needing a recital of what was in the original.

I have spent a lot of time in the past reading articles for the purposes of a review article, and it is amazing how many papers are cited which merely refer to a further article, rather than the original source. None of them added anything to the knowledge of the subject. What is rewarding is when you eventually track down the original work and see what those workers did.

It is much like the acceptable way of quoting from a post to which you are responding. A reference or link and a brief summary of the point is all that is needed.

TJH

9873210
Lemon Slice
Posts: 986
Joined: December 9th, 2016, 6:44 am
Has thanked: 226 times
Been thanked: 297 times

Re: The Thorny Issue of Fraudulent Research

#48655

Postby 9873210 » April 25th, 2017, 8:31 pm

After suffered through strings of least publishable units I could only wish somebody would indulge in a spot of plagiarism and sum up the whole chain every 20 links or so.

Injunear
Lemon Pip
Posts: 65
Joined: November 9th, 2016, 9:25 pm

Re: The Thorny Issue of Fraudulent Research

#52338

Postby Injunear » May 9th, 2017, 6:34 pm

Anther area of conspicuous fakery is the promotion of the "lipid hypothesis" about the causes of cardiovascular diseases.


To be fair to the Guardian I believe they have today published a story that rather debunks the lipid hypotheses, or at least the consumption of dairy component.

(I know - it's unlike me to be fair about the Guardian, but they mostly bring it on themselves).

Lootman
The full Lemon
Posts: 18681
Joined: November 4th, 2016, 3:58 pm
Has thanked: 628 times
Been thanked: 6564 times

Re: The Thorny Issue of Fraudulent Research

#52350

Postby Lootman » May 9th, 2017, 7:06 pm

Last year's film "Denial" with Rachel Weisz, Timothy Spall and Tom Wilkinson, covers an issue that overlaps with this - academic research purporting to deny the Holocust. And shows that this problem is hardly a new one.

XFool
The full Lemon
Posts: 12636
Joined: November 8th, 2016, 7:21 pm
Been thanked: 2608 times

Re: The Thorny Issue of Fraudulent Research

#52361

Postby XFool » May 9th, 2017, 8:19 pm

Injunear wrote:
Anther area of conspicuous fakery is the promotion of the "lipid hypothesis" about the causes of cardiovascular diseases.

To be fair to the Guardian I believe they have today published a story that rather debunks the lipid hypotheses, or at least the consumption of dairy component.

To be fair to the quoted text, here is it's fuller context:

It's a bit rich for the Guardian to carry that, given its devotion to the fakery of "Climate Science".

Anther area of conspicuous fakery is the promotion of the "lipid hypothesis" about the causes of cardiovascular diseases.


Return to “Science”

Who is online

Users browsing this forum: No registered users and 7 guests