The pressures of publish or perish
Mac prof ’s retraction case highlights issues that can impact scientific ethics
There are a lot of questions still to be answered in the case of Jonathan Pruitt, a McMaster University professor and Canada 150 Research Chair now facing intense scrutiny in the scientific community after two of his publications were recently retracted by academic journals.
Another dozen or so retractions may still be coming after Pruitt’s co-authors found serious problems with his data, which included numbers that were duplicated or even repeated in larger blocks.
Was it a case of honest mistakes? Sloppiness? Was the data falsified or even fabricated? Or is there a logical scientific explanation for the data?
And if it’s shown the data were manipulated, how did it happen? Who was involved? How far back does it go? How many careers will be damaged?
Many of the answers will have to wait. Pruitt is reportedly in the midst of a four-month research project in Australia and the South Pacific and doesn’t appear to have made any public comments about the controversy since a late-January interview with Science Magazine.
But Pruitt’s unfolding case shines a light on how a collision of several factors can put the squeeze on scientific ethics — the pressure on researchers to publish academic papers, the pressure to obtain funding, and what’s known as publication bias.
In science, the phrase “publish or perish” is well known.
Publications in academic journals are often used as the measuring stick for researchers, be they professors, postdoctoral fellows or graduate students. Obtaining a job, a promotion, or getting tenure can depend heavily on the quantity and quality of a researcher’s publications.
“The job market is incredibly competitive and young academics must publish or they will not obtain positions,” said William Hughes, a Canadian post-doctoral researcher in evolutionary biology at Stockholm University. “It is that simple.”
“I believe that pressure to publish is probably the main factor in encouraging unethical behaviour,” said Hughes, who is from Ottawa and obtained his PhD from Carleton University.
As a peer reviewer, Hughes said it’s not uncommon for him to see plagiarized papers, “even though it is both highly unethical and very easy to detect.”
“I suspect that these authors only take such a risk because they are desperate to publish,” Hughes said.
Tied in with the need to publish is the need to obtain funding.
Science is rarely done alone. A university lab is made up of professors, post-docs, graduate and undergraduate students, technicians and equipment. That can require a lot of money.
The competition for a limited pot of grant money can be fierce. Decisions on who wins can be influenced by the publication records of the competitors.
Not all academic publications are created equal either. There are hundreds of journals but they exist in a hierarchy, with Nature and Science at the top of the pyramid, and researchers face pressure to publish in the best journals possible.
“Publishing is therefore a track record of previous scientific success that can be leveraged to obtain funding,” said Hughes.
And finally, there’s the phenomenon known as publication bias.
Studies have shown that academic journals are predisposed to publish results that have positive findings rather than neutral or negative results. In other words, there’s a bias toward publishing studies where someone had an idea, tested it and found they were right.
“It’s human nature that it’s more interesting to get a new result than to say ‘I tried this experiment and not much happened,’ ” said John Berlinsky, a retired McMaster professor and former chair of the department of physics and astronomy.
But publication bias can be harmful. First, it can influence researchers — subtly or not so subtly — to make sure their experiments show something positive because they know those results have a better chance of being published.
Second, it can be as important in science to know what doesn’t work so people don’t waste their time trying. But if those results don’t get published, it stays in the dark.
Tying it altogether, it means researchers can be under pressure to publish for career reasons, they’re under pressure to publish in high-quality publications, they’re under pressure to come up with ideas that lead to positive results, and their publication record can have an impact on how much money they receive.
Nonetheless, Berlinsky doesn’t believe there’s widespread research fraud.
“It is actually difficult to fabricate,” said Berlinsky. “It’s not very easy to get away with it because if anyone cares about it, people will check it.”
The primary motivation for most scientists, Berlinsky said, is peer recognition.
“If someone is actually motivated by peer recognition, then the last thing they want is to do something that will basically end their career,” said Berlinsky. “Secondly, it’s hard to do it in a way you’ll get away with it. And thirdly, people will find out anyway.”
“So yes,” he said, “there are pressures to produce exciting results but the actual economics of fabricating results doesn’t really work in favour of the scientist.”