Half of academics admit to ‘questionable’ research practices

Findings from Dutch confessional covering methods, supervision and publishing likely to be repeated in other countries, authors say

March 27, 2022
Confession booth

Academics who cut corners for personal gain can count themselves in the majority, according to findings from one of the largest surveys of academic integrity worldwide.

Overall, 51.3 per cent of respondents to the Dutch National Survey on Research Integrity frequently engaged in at least one questionable research practice. “This is, you could say, almost common,” said Lex Bouter, a professor of methodology and integrity at Amsterdam University Medical Centers, who co-led the survey.

Respondents to the anonymised questionnaire could state their practice of 11 questionable acts, ranging from not publishing negative findings to fabricating data. The most common confessions were not submitting negative studies for publication, playing down study limitations, under-supervision of junior colleagues, neglecting equipment, skills or expertise, and inadequate note-taking. 

“Questionable research practices might on the aggregate, because they occur so often, have a larger impact,” said Professor Bouter, comparing them to scientific misconduct. “They’re more preventable in the sense of talking about them and teaching about them: I cannot envision a course preventing data fabrication, but I do envision a course that prevents selective reporting and p-hacking.” 

The authors found being male and junior increased the likelihood of questionable behaviour. They found that questionable practices were most common among biomedical researchers, at 55.3 per cent, and that this group also topped the table for fabricating and falsifying results, at 10.4 per cent. 

Professor Bouter cautioned against taking too much from disciplinary differences in questionable practices. “They might partly mean different things in different fields,” he said.

“My expectation would be that if we would repeat it in other countries, we would basically find similar results,” said Gerben ter Riet, an author and professor of epidemiology at Amsterdam University Medical Centers, adding that “local research cultures, and recognition and reward systems” could influence behaviour. 

Gowri Gopalakrishna, a co-leader of the survey and a postdoctoral researcher at Amsterdam University Medical Centers, said incentive and reward structures should recognise mentoring and supervision, the neglect of which was the third most common misdemeanour. “Mentoring can really go both ways,” she said, contrasting the “open and collaborative” habits learned from responsible mentoring with what the authors call ‘survival mentoring’, which is “more about teaching you to cut corners”. 

The overall findings of the survey, based on 6,813 responses, were published in the journal Plos One.


Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.

Related articles

As the pandemic increases public scrutiny of science, the UK Parliament is holding another inquiry into the long-running issue of reproducibility. Five of its contributors give their views on how sloppy science can be eliminated and trust be more firmly rooted 

20 January