Every researcher in the Netherlands is to be questioned about whether they have committed research misconduct or engaged in “sloppy science” as part of a major national effort to bolster scientific standards.
In response to rising concerns over a “reproducibility crisis” in science and a series of high-profile fraud cases in the Netherlands, the country is to commit €8 million (£6.3 million) to understanding the problem, finding solutions and trying to reproduce critical studies.
Lex Bouter, professor of methodology and integrity at VU Amsterdam and one of the driving forces behind the initiative, said that in the Netherlands “during the past 10 years we had three to four really serious wake-up calls” about scientific misconduct.
The most prominent is the case of Diederik Stapel, who was sacked by Tilburg University in 2011 for fabricating data, and who left a trail of fraudulent social psychology papers.
Professor Bouter started pushing for a major initiative to investigate research misconduct during a sabbatical year when he began to focus on research integrity. “The evidence is so poor in these areas that the funding councils…decided: ‘Let’s take some action’,” he explained.
There are two parts of the Dutch investigation into research integrity: a €5 million programme called Fostering Responsible Research Practices, which will include a €1.2 million national survey of researchers and €2.7 million of research grants into the area; and a €3 million pot to fund replication studies of important “cornerstone” research that has been relied on to make policy or has attracted lots of media attention.
The programmes are set to be signed off soon and calls for proposals are expected before the end of the summer.
The mass survey of researchers in the Netherlands will encompass all disciplines including humanities scholars, who, like scientists, can be “selective” in their use of sources, he said. The anonymised results will be reported back to universities so they can judge the extent of the problem.
A number of researchers have already attempted to estimate the extent of misconduct. A 2009 meta-study found that 2 per cent of scientists admitted to having “fabricated, falsified or modified data or results”, although Professor Bouter said that this was likely to be a “gross underestimate”.
He argued that there was still “not that much evidence” on the subject. “When you want to be really effective and efficient in your intervention, you need to know what works,” he said.
Three years ago, when Professor Bouter talked to scientists about misconduct, “there were often hostile reactions from the audience”. But this is no longer the case, he said.
