A survey, completed by 6,813 scientists from 14 Dutch universities and university medical centers, found that 51.3 percent of those surveyed are sometimes guilty of questionable research practices.
Concretely, this includes not publishing research if results are disappointing, failing to report publication deficiencies, or reporting only desired results from statistical tests. The greatest harm in such practices lies in the fact that important information is omitted, according to psychologist Geert Storms (KU Leuven).
The most frequently cited reason is the pressure to publish within the research culture. Storms says that there is “a kind of Darwinian process in science,” in which a lack of publications equates to a lockout of money.
This was also endorsed by Professor of Methodology Lex Bouter (VU Amsterdam), who was involved in the research. According to him, pressure to publish could have consequences for the validity and trust of science. Either from the community or among the scholars themselves.
almost important
The same problem emerged in a recent study by psychiatrist and researcher Joeri Tijdink (VU Amsterdam) and colleagues. They analyzed trends in the ways researchers describe results that are not statistically significant.
The word choices used are often “almost significant,” “significant trend,” and “positive trend.” By presenting the results more positively than they actually are, the chance of them being published increases.
“Science is a very big ship, and it is not easy to change course,” Storms says.
However, there are possibilities. Microbiologist Elizabeth Beck, a science integrity advisor at Harbers Bik LLC, has exposed several questionable practices in recent years.
According to her, control is an important point. Withdrawing papers can be an effective warning strategy. I also found it important that research institutes do not cover up malpractice.
As far as Tijdink is concerned, there has to be a change in the way scholars are recognized and evaluated. “Instead of counting posts, we should look more at the quality of the world as a person and the quality of his or her work.” Tijdink advocates “slow science,” whereby more time is available for comprehensive reflection through research.
Bouter also sees greater benefit in rewarding “good behavior” than in promoting a culture of control. “Scientists are human too, and you have to provide the right incentives if you want them to behave well.”
Another way to counter questionable research practices, Tijdink says, is by introducing a two-step publishing process. Researchers submit an introduction and research method to a professional journal before submitting the final paper. The journal then determines in advance whether the research will be published regardless of the outcome. “If the methods are logged correctly before the research is done, you’ll get more honest and reliable results,” Tejdink says.
multi-headed beast
After the so-called Stapel Affair in 2011, which featured years of scientific fraud by social psychologist Diederik Stapel, changes were promised. What about that? According to Storms, “something is moving, but not enough.”
Bouter also says that changes were certainly made after the Stapel case. He points out that shady practices may not be more common today than they were in the past, but we are more aware these days that they are happening. The problem of questionable practices in research is a multi-headed beast and it has not yet been solved.
