The journal NATURE reveals peer-reviewing scam

Post Reply New Topic

As an undergraduate student in music history, my academic advisor, who, in retrospect, may have been on the spectrum, was always complaining about how most modern research was bogus. He was always also complaint about how there was so much "slicing and dicing" going on so that researcher, as well as their institutions did it to keep the federal, state, and private funds coming in.

And this, published last week:

http://www.theguardian.com/science/head ... s-lancioni

Somewhat surprising that a journal considered respectable (until this) had a one day time lapse between submission and acceptance of articles. Somewhat disheartening that some academics turned on the whistleblower, not the culprit.

I saw it happen close up as an undergraduate (by an associate professor who stole data that students had done and published it under his own name).

I also remember one master's student who stole his supervisor's work and claimed later that the supervisor had stolen it off him in the first place...

I remember a behaviorist professor who rudely threatened to fail the exam papers of any student who criticised behaviourism as a theory (he wasn't joking).

There were no consequences that I ever heard of, these matters were swept under the carpet, dealt with 'in camera' if at all, and it was business as usual. The reason for this though may have been that it was a sacking offence for any member of a faculty to say or publish something that "would be likely to bring the university into disrepute". So the senate itself was implicit in cover-ups, because that provision silenced whistleblowers.

Perhaps those early experiences are why I have never bought in to the "romance" of pure unsullied science, and I don't believe every study I read just because it's from a big name university and "peer reviewed" or "evidence based". (There are lots and lots of ways to treat data to get the result you want, without actually faking raw data. Instead you only report the proportion of total tests that show what you want to show). These are much much harder to detect, and I would guess that 98% are never detected. Journals hardly ever demand all the raw data, so the fakers feel safe.

Sometimes peer review only means that one cheat reviewed another cheat - they certainly are peers on that score when this happens. And a prestigious journal is no guarantee either, as these links have demonstrated.