Research Misconduct

Author: Michael Kalichman, 2001
Contributors: P.D. Magnus, Dena Plemmons
  • Necessity for self-policing
    The details of how research is conducted are often known only to those actually working on a project. This relative secrecy is driven by many different factors, from sheer practicality, to protection of credit or intellectual property rights, to worries about the possible misuse of preliminary data. Where there is this secrecy, however, misconduct will only come to light if someone close to the project blows the whistle.
  • Obligations to act
    Scientists do not all agree regarding if, when, and how to report misconduct. There is a considerable range of opinions about how to respond to perceived misconduct -- and an even greater difference between scientists and administrators (Wenger et al., 1999). Yet, as a 1995 publication of the National Academy of Sciences advises, 'someone who has witnessed misconduct has an unmistakable obligation to act.'
  • Questions about research
    Questions about the proper conduct of research are frequent and to be encouraged, but only rarely is the problem about research misconduct. It is a responsibility of all scientists to find the best and most appropriate means to address concerns about the conduct of science.

Science is predicated on trust. Without confidence in the integrity of their peers, scientists would be unable to trust one another's work. Even if the demands of ethical and responsible conduct may not always seem expedient, they are always necessary to the enterprise of science. Self-regulation and self-policing operate to ensure the legitimacy of research, and necessitate that scientists foster an environment in which responsible research is explicitly discussed and encouraged. In part, this means that scientists should be familiar with definitions of research misconduct and procedures for dealing with it, regardless of whether they will ever be party to allegations.

How frequently does research misconduct occur? There are some indications that questionable research practices may be common (e.g., Kalichman and Friedman, 1992; Martinson et al., 2006), but that research misconduct occurs only rarely. In 20 years, the federal government found an average of about 10 cases of research misconduct per year; that is, about 1 case per year for every 100,000 researchers. However, there are many barriers to accurately quantifying the extent of research misconduct; cases may go unreported and institutions may be biased against finding misconduct. The actual rate of research misconduct could be as low as 1 in 100,000 or as high as 1 in 100 (Steneck, 2000; Steneck, 2006). Yet, in the past 25 years, many serious allegations of misconduct have been widely publicized, and some of those were borne out by subsequent investigation.

A government-wide definition of Research Misconduct was proposed by the Office of Science and Technology Policy (OSTP, 2000) and is now covered in the Code of Federal Regulations for both the Public Health Service (PHS, 2006), the National Science Foundation (NSF, 2006), and other agencies as well. In all cases, research misconduct is essentially defined as: "fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results."

Minimally, for something to count as research misconduct it must be committed intentionally, knowingly, or recklessly, and there must be a significant departure from accepted practices of the relevant research community. Not all instances of misbehavior or questionable conduct are covered under these policies, but for those practices that are covered, there are explicit steps that must be taken in the event of an allegation of misconduct.

Case Studies

Discussion Questions

  1. Define fabrication, falsification, and plagiarism.
  2. Give at least three examples of misconduct by researchers that would not meet the existing or proposed definitions of research misconduct. In your institution, what can be done about these types of misconduct?
  3. In your institution, what formal procedures or mechanisms (e.g., ombudsman, conflict resolution, arbitration, mediation) are available to help resolve disputes or questions about the responsible practice of science?
  4. Outline the basic steps to be followed in your institution for responding to an allegation of research misconduct.
  5. If you have direct evidence that someone in your institution has committed research misconduct, then to whom and how should such an allegation be made?
  6. If you were accused of having fabricated data that you had produced, how could you demonstrate that you have actually obtained the results you reported?

Additional Considerations

  • Obligation to act
    Most people will find it difficult to be silent about wrongdoing, particularly if a failure to speak up is likely to result in harm to patients or subjects, a waste of scarce resources, or publication of misleading findings. Others may be inclined to report misconduct because they would not want to risk that an independent discovery of the misconduct could implicate them for complicity or could at least lead to questions about why nothing had been said earlier. Finally, the public and other funders of research have the right to expect that recipients of the funding will address serious deviations from good research practice.
  • Rules and procedures
    Although institutions receiving federal funds need to meet a common set of minimal requirements, individual institutions are granted substantial leeway in the rules and procedures for handling of allegations of misconduct. Especially if you become involved in an allegation of misconduct, it is in your best interest to familiarize yourself with all relevant institutional procedures.
  • Documentation
    An allegation of research misconduct is one of the most serious charges that can be made against a scientist. As such, it is essential that a charge be sustained only if justified by documentation and other relevant evidence. Whether one is making the allegation or accused of misconduct, clear documentation provides the best chance for a fair and timely resolution. As with good research, an allegation of misconduct should be sustained or rejected based on adequate documentation.
  • Public allegations
    The pace of the process for dealing with alleged misconduct may be frustrating. In such circumstances, it can be tempting to discuss the case publicly. However, placing a complex, unresolved issue into the public arena can produce unpredictable results, which can be harmful to those directly involved and to the scientific community as a whole. Publicity may also compromise the integrity of an ongoing inquiry and the privacy of parties to the investigation. Moreover, an attempt to circumvent the institutional process may prejudice those charged with reviewing the allegation.
  • Not necessarily research misconduct
    Even when a strong argument can be made for action, making an allegation of research misconduct should not be a first step to remedy questions or concerns. Some aspects of conduct are too new or poorly defined to allow for a simple answer about what is appropriate. Other behaviors may stem from bad manners, honest error, or differences of opinion, which may be questionable without being research misconduct. Impressions should be validated before making serious charges, and many apparent problems can be resolved by other means.
  • Dispute resolution
    Many concerns are best addressed by means other than alleging research misconduct. Some institutions have formal mechanisms in place for conflict resolution, mediation, or arbitration; absent such mechanisms, finding a solution to a dispute may require some creativity.
    • Conflict resolution: Often, good conflict resolution skills may be helpful or even sufficient. Deal with the problem as early as possible. Begin by defining points of agreement and then work on areas of disagreement. Emphasize the problem rather than the person. Give and ask for clear communication about what is most important to each of the interested parties.
    • Mediation: A respected third party can sometimes help with mediating a dispute. The goal is to clarify issues in a way that permits the best possible agreement or compromise.
    • Arbitration: When other avenues of communication have failed, then parties to a dispute might be convinced to put their cases before a mutually agreeable arbitrator for review and a binding decision.
  • Kalichman MW, Friedman PJ (1992): A pilot study of biomedical trainees' perceptions concerning research ethics. Academic Medicine 67: 769-775.
  • Martinson BC, Anderson MS, Crain AL, de Vries R (2006): Scientists' Perceptions of Organizational Justice and Self-Reported Misbehaviors. Journal of Empirical Research on Human Research Ethics 1:51-66
  • NSF (2005): Sec. 689.1 Definitions. Part 689-- Research Misconduct. Subpart A—General. Chapter VI--National Science Foundation. Title 45--Public Welfare. 45CFR689.1(a).
  • OSTP (2000): Federal Policy on Research Misconduct: Notification of Final Policy. Federal Register December 6, 2000 65(235): 76260-76264.
  • PHS (2005): Sec. 93.103 Research misconduct. Part 93-- Public Health Service Policies on Research Misconduct. Subpart A—General. Chapter I--Public Health Service, Department of Health and Human Services. Title 42--Public Health. 42CFR93.103.
  • Steneck N (2000): Assessing the integrity of publicly funded research: A background report for the November 2000 ORI Research Conference on Research Integrity.
  • Steneck N (2006): Fostering Integrity in Research: Definitions, Current Knowledge, and Future Directions. Science and Engineering Ethics (2006)12: 53-74.
  • Wenger NS, Korenman SG, Berk R, Honghu L (1999): Reporting unethical research behavior. Evaluation Review 23: 553-570.