Addressing Research Fraud

The Good Science Project is launching a program to promote investigations into research fraud and other serious misconduct.

We are uniquely poised for such a program: The Good Science Project is run by Stuart Buck, who (as an Arnold Foundation program officer) was responsible for funding many initiatives in open science and reproducibility, and who graduated with honors from Harvard Law School, was on the Harvard Law Review, clerked for a DC Circuit judge, and worked at a top DC law firm.

Our goal is to enable more people to expose scientific fraud. We will do so by engaging in the following tactics:

• Provide education and advice to people who want to blow the whistle on potential fraud or other serious research misconduct;

• Recruit an expert advisory board to help make sure that whoever we help is both legally and scientifically defensible;

• Find and organize private lawyers (including law firms and organizations such as the ACLU) who are willing to provide legal advice, pre-publication review, representation to people accused of defamation, and other services, such as referring to existing legal resources, including the Scientific Integrity Fund.

The overall program is explicitly modeled after the long-time success of the Climate Science Legal Defense Fund. Indeed, we can build on their pre-existing efforts. For example, CSLDF produces a major report on “research protections in state open records laws” as to all 50 states plus DC, as well as 12 guides to the scientific integrity policies at all major federal agencies that fund research.

We are starting with a focus on social science and medicine, since those are the areas where we already have experience. As well, those are areas with substantial impact as to public policy and health.

Bottom line: If you know of possible scientific fraud in those fields, and are nervous about publicly discussing it or being a whistleblower, please reach out to us.

The Problem

Research fraud is far too prevalent in many fields. Just recently, the National Institutes of Health (NIH) issued a finding that Dr. Eliezer Masliah (who had led a $2.6 billion program on Alzheimer's disease) was guilty of "falsification and/or fabrication" in his own published articles. As reported in Science, Masliah apparently has a long string of questionable articles stretching back to the 1990s. No one noticed until now.

Masliah isn't an isolated anecdote, but is only the latest example of a much broader phenomenon.

In one of the best articles on this subject, Daniele Fanelli examined a total of 21 surveys, and found that on average, nearly 2% of scientists “admitted” to having “fabricated, falsified, or modified data or results.” Self-admissions are obviously a huge understatement, since most people who commit fraud would never admit it even on an anonymous survey. When scientists were asked about the behavior of their colleagues, fully 14% of scientists were aware of falsification and up to 72% had seen “other questionable research practices.”

A more recent meta-analysis of surveys found that 2.9% of scientists admitted to research misconduct (defined to include falsification, fabrication, and plagiarism), while 15.5% had “witnessed others” who committed research misconduct. Moreover, the Dutch National Survey on Research Integrity (with some 6,800 respondents, more than any previous survey) found the following, as reported in Science:

More than half of Dutch scientists regularly engage in questionable research practices, such as hiding flaws in their research design or selectively citing literature, according to a new study. And one in 12 admitted to committing a more serious form of research misconduct within the past 3 years: the fabrication or falsification of research results.

A one-in-12 rate is quite high, and even this is likely to be an understatement, as it relies on self-reports. More recently, James Heathers has estimated that the prevalence of scientific fraud might be as high as 1 in 7 scientific papers.

In short, serious misconduct and outright fraud are likely more common than is comfortable to contemplate. All the more reason that we should nurture and protect folks who are willing to call out problems (where justified).

Notwithstanding the above, raising methodological concerns about other people’s research is often unpopular. If you want to spend time digging into someone else’s scholarly work to investigate potential fraud, you will find that:

• there are few-to-no grants;

• there are few-to-no possibilities for getting a top publication; and,

• no university advertises a faculty position in “Calling Out Fraud In Our Other Scholars.”

Small wonder that graduate students and early-career researchers who are potential whistleblowers are often warned by their own mentors not to waste time chasing down fraud in someone else’s research, and told that it is infinitely better to spend your time on publishing your own research.

Our advisory board includes:

• Glenn Begley

• Lee Ellis

• Chelsea Polis

• Simine Vazire

• Elisabeth Bik

• Dorothy Bishop

We also have a working relationship with John Knepper, a lawyer with many years of experience in state and federal government (including Treasury, DOJ, and the White House), and who is available to help scientists serve as whistleblowers for potential fraud.

In the long run, we hope to change scientific incentives. In the meantime, we want people who see signs of scientific fraud to be empowered with social support and legal advice.