The image of scientists as objective seekers of truth is periodically jeopardized by the discovery of a major scientific fraud.
How frequent are scientific frauds?
A popular view propagated by the media and by many scientists sees fraudsters as just a ‘‘few bad apples’’.
Increasing evidence, however, suggests that known frauds are just the ‘‘tip of the iceberg’’, and that many cases are never discovered.
What constitutes scientific misconduct?
Different definitions are adopted by different institutions, but they all agree that fabrication (invention of data or cases), falsification (wilful distortion of data or results) and plagiarism (copying of ideas, data, or words without attribution) are serious forms of scientific misconduct.
[R]outine data audits conducted by the US Food and Drug Administration between 1977 and 1990 found deficiencies and flaws in 10–20% of studies, and led to 2% of clinical investigators being judged guilty of serious scientific misconduct.
[D]ata fabrication and falsification are rarely reported by whistleblowers, and are very hard to detect in the data.
Even when detected, misconduct is hard to prove, because the accused scientists could claim to have committed an innocent mistake.
Distinguishing intentional bias from error is obviously difficult, particularly when the falsification has been subtle, or the original data destroyed. In many cases, therefore, only researchers know if they or their colleagues have wilfully distorted their data.
This is the first meta-analysis of surveys asking scientists about their experiences of misconduct.
It found that, on average, about 2% of scientists admitted to have fabricated, falsified or modified data or results at least once –a serious form of misconduct my any standard and up to one third admitted a variety of other questionable research practices including ‘‘dropping data points based on a gut feeling’’, and ‘‘changing the design, methodology or results of a study in response to pressures from a funding source’’.
In surveys asking about the behaviour of colleagues, fabrication, falsification and modification had been observed, on average, by over 14% of respondents, and other questionable practices by up to 72%.
[W]hile surveys asking about colleagues are hard to interpret conclusively, self-reports systematically underestimate the real frequency of scientific misconduct.
Therefore, it can be safely concluded that data fabrication and falsification –let alone other questionable research practices- are more prevalent than most previous estimates have suggested.
Indeed, there seems to be a large discrepancy between what researchers are willing to do and what they admit in a survey.
In a sample of postdoctoral fellows at the University of California San Francisco, USA, only 3.4% said they had modified data in the past, but 17% said they were ‘‘willing to select or omit data to improve their results’’.
Among research trainees in biomedical sciences at the University of California San Diego, 4.9% said they had modified research results in the past, but 81% were ‘‘willing to select, omit or fabricate data to win a grant or publish a paper’’.
Overall, admission rates are consistent with the highest estimates of misconduct obtained using other sources of data, in particular FDA data audits. However, it is likely that, if on average 2% of scientists admit to have falsified research at least once and up to 34% admit other questionable research practices, the actual frequencies of misconduct could be higher than this.