Spinning scientific results: ‘Tempting, but never justified’

Spinning facts is not just something politicians and lawyers do. Scientists, too, hype their results and give a rosy description of their findings. Is ‘spin’ ever justified? According to PhD student Michèle Nuijten, the answer is no: “Exaggerating your findings can be tempting, but it can also be very harmful.”

Groundbreaking discoveries, breakthroughs and cures: biomedical science is thriving. Or is it not? According to a new Australian study, a large number of clinical trials published in medical journals are sensationalized. The authors of the study conclude that a majority of biomedical papers contain ‘spin’, meaning that findings are presented in a way that makes them look more favorable than they actually are.

Gray area

Michèle Nuijten, PhD researcher at Tilburg University’s department of Methodology and Statistics, believes that spin is not just prevalent in biomedical literature. “It happens in other fields too”, she says. “Even from my own experience, I know it can be tempting. As a scientist, you have to interpret your data in order to connect meaning to your findings. When you write a paper, it can be difficult to present your findings in clear manner, while also being very accurate and precise.”

Although Nuijten understands why research papers are vulnerable to spin, she warns that spin can make science misleading. “If researchers are convinced of the importance of their study, they may be inclined to interpret results in an overly positive manner. They may leave out non-significant results, remove outliers from their analysis, or confuse correlation with causation to give a more rosy description of their findings.”

According to Nuijten, twisting results to support exaggerated conclusions does not amount to fraud. “Committing fraud means that you fabricate data. Spin falls into a much larger, gray area, making it a difficult problem to tackle”, she says. “In my opinion, twisting results is never justifiable. It can be harmful to science, and to society as a whole.”

Exaggerated exaggerations

Although researchers who twist results may not be acting with the intent to provide false information, Nuijten warns that it can have serious effects. “If you exaggerate your findings, you create a snowball effect. If the press reports on those already overstated findings, your research findings will be simplified and exaggerated further. And, subsequently, the audience who ends up reading about your research will draw even less accurate conclusions from what is reported in the media.”

“Exaggerated conclusions can be harmful. I recently came across a ‘science-based’ cookbook with recipes to fight cancer”

Sometimes, exaggerated conclusions in scientific papers have a serious impact on society, Nuijten says. “We often see this snowball effect in food science: researchers publish an article about gluten, the media report on it, and suddenly those findings are blown completely out of proportion. And that’s just gluten – there are more serious examples. I recently came across a ‘science-based’ cookbook with recipes to fight cancer.”

Sensationalist culture

According to Nuijten, it’s not just the researchers themselves who are to blame for the increasing prevalence of spin in scientific papers. “It’s a complex problem, to which more than one factor contributes. On the one hand, there’s human error and the temptation of interpreting your results in a way that is favorable. And at the same time, the sensationalist culture of science doesn’t help.”

“Science is a process, in which groundbreaking discoveries are scarce”

Papers are often only considered relevant if they present groundbreaking findings, Nuijten says. Spectacular conclusions seem to be more important than solid statistics. “I recently peer-reviewed a paper for Proceedings of the National Academy of Sciences, or PNAS, which is a reputable scientific journal. Like other top journals, PNAS has a policy that places the methods section at the very bottom of the article. It’s even printed in a smaller font than the rest of the paper. As if it were just a footnote!”

According to Nuijten, science should be less concerned with spectacular conclusions, and more with nuance. Breakthroughs are scarce – or at least, real ones are. “Science is a process. If we would all be a little bit more accurate, it would be easier to build on each other’s work and really move forward.”

_Advertentie.