About scientific misconduct

Here are a few starters to think about scientific misconduct issues:

Note about the authors of the papers listed on this website

Disclosure of the names of the authors of papers listed on this website raises some debate: isn't it like a double penalty for misbehaving authors who have already been punished for their misconduct? How about innocent co-authors? One cannot argue co-authors are responsible for the entire content of the papers, especially when data are collected or analyzed separately, as is the case in collaborations. On the other hand, it seems difficult to hide the names of the authors, since this information remains available on the publication itself, on the journal, and on any online database.

Therefore, we have decided not to hide the authors' names, but not to mention the person responsible for misconduct (the name of this person is only available on some assessments of misconduct which are linked to the publications).

Most importantly, it should be kept in mind that misconduct is always context-dependent: the fact that one once misbehaved does not mean he/she will never again be able to behave properly and produce reliable scientific results!

A definition of scientific misconduct?

One of the difficulties to address the issue of scientific misconduct at the international level stems from the wide diversity of definitions and procedures among countries and institutions. For instance, the US Office of research integrity has adopted “fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results.” as a definition. This definition might turn out to be too restrictive, overlooking other misbehaviours such as ignoring ethical human-subjects requirements. The Committee on publication ethics has preferred a broader definition: “behaviour by a researcher, intentional or not, that falls short of good ethical and scientific standards”.

Alternatives to defining misconduct include attempts to take inventory of all misbehaviours that undermine science integrity (e.g. see this unofficial report by the OECD), or proposals for a code of conduct, as described below.

How prevalent is scientific misconduct ?

Few studies have estimated the prevalence of misconduct amongst researchers. In a survey published in Nature, US scientists were asked to report their own behaviour (anonymously). Overall, 33% of the respondents said they had engaged misbehaviour in the past three years. Misbehaviours ranged from “Falsifying or ‘cooking’ research data” (0.3%) to “inappropriately assigning authorship credit” (10%) or “Changing the design, methodology or results of a study in response to pressure from a funding source “ (15.5%) for instance. These observations led the authors to argue that “To protect the integrity of science, we must look beyond falsification, fabrication and plagiarism, to a wider range of questionable research practices”.

Other studies have focused on misconducts arising in the context of conflicts of interests, in biomedical or pharmaceutical research. Although conflicts of interest arising from financial relationships between industry, scientific investigators, and academic institutions can influence biomedical research in important ways, it is worth stressing that conflict of interest is not synonymous of falsified data. We think that authors should not fear to disclose conflicts of interest if they feel that research has been properly conducted.

Scientific misconduct: a taboo issue?

At different stages, scientific misconducts tend to be overlooked: misconduct goes underreported, leaders of institutions are reluctant to launch investigations, journals sometimes do not retract erroneous papers.

A study published in Nature estimates that 3 incidences per year are observed per 100 researchers, whereas only 24 investigations are submitted to the Office of Research Integrity every year. According to the respondents, “58% of the observed incidents had been reported to officials at their institutions.”

On the other hand, as mentioned in another paper in Science, leaders of institutions have their share of responsibility. They are sometimes reluctant to conduct investigations, since misconduct is likely to jeopardize the name of their institution. This can seem to be paradoxical since such investigations could show how the institution is able to reliably handle misconduct. Investigations can also be costly in terms of time and money, and sometimes institutions are poorly equipped to conduct them.

Finally, journals also have a role in disclosing information about the integrity of papers. When fraud is assessed, some journals do not publish any note nor retract the paper, sometimes because they require that all the authors agree with the retraction.

Retracted papers keep being cited !

“If people cite fraudulent articles, then either their research is going to be thrown off or something will be wasted,” said Paul Friedman, a former dean at the University of California. However, a note published in Science stresses how uneasy it is to keep updated about retracted papers. Online retractions don’t always get picked up, or retractions often occur after people have already downloaded the paper.

However, publicity might have a key role in cleaning retracted papers off the literature: it has been observed that “the less media coverage [a retracted paper] receives, the more likely it [is] to continue to be cited.”

A code of conduct?

Several bodies, such as law, engineering, or medicine already have code of conduct. What about such a code for scientists? Dr. Stephen Unger, former chair of the Institute of Electrical and Electronics Engineers (IEEE) Ethics Committee, argues that a code of ethics can help create an environment in which ethical behaviour is the norm, and can serve as a guide.

Some scientific institutions already have their code of ethics, available on the web. The Student Pugwash USA also initiated a pledge program. For J. Rotblat, “scientists can no longer claim that their work has nothing to do with the welfare of the individual or with state policies.” He emphasizes the need for scientists to be aware of their responsibilities, not only towards science but also towards the society.

Causes of scientific misconduct

A committee (formed by the IOM and the NRC) addressed the issue of the influence of the research environment. For instance, pressure for funds, for publication, for achieving positive results (especially in the case of sponsored research) are frequent in the life of a researcher. In addition, researchers are often not aware of the rules of scientific conduct, and of the punishment someone found guilty of misconduct can get. The committee also emphasized the need for an appropriate and creative education in the responsible conduct of research.

Education

To build a culture of research integrity, proper training of current and future researchers is essential. A researcher is less likely to "misbehave" when they have internalized rules of conduct, but also when they evolve in a sane environment where research misconduct is not overlooked.

Most trainings in ethics focus on "bioethics", addressing specific issues raised by new technologies or experiments with humans and animals for instance. Trainings that aim at raising awareness about scientific integrity and develop skills to handle misconduct exist in some universities, but are far from being common.

In a book entitled “Integrity in Scientific Research: Creating an Environment That Promotes Responsible Conduct”, the authors claim that “the responsible conduct of research is not distinct from research”. They argue that integrity should be developed in an overall research education program, in a creative way. The Responsible conduct of research education committee also provides lots of online resources for promoting such education programs.

Available information about research integrity in publications

Publications constitute:

  • a report and a record of the activity of researchers;
  • the references research builds upon;
  • the data governments or society can refer to when facing issues that have a major scientific component (public health, environment...);
  • a way of evaluating the activity of individual researchers.

The quality of publications, in terms of scientific integrity, is therefore essential for research to be conducted in an efficient and responsible way, and for a transparent communication between researchers and society.

More importantly, access to information concerning the integrity of publications is crucial. This would help readers to identify dubious publications, and show how the scientific community is able to self-criticize in order to build as reliable knowledge as possible.

Debates about specific kinds of misconduct