Group of Eight Australia
Australia's Leading Universities
     Modules     Organiser  

Module 3: Conducting Research Responsibly – Protecting Yourself, Your Research, and Your University

3.5 Research misconduct and questionable research practices

Just as research integrity sustains a reputation hierarchy, the other side of the coin is that research misconduct and questionable research practices can bring this into jeopardy. One high profile international example that jeopardised this reputation was the case of Andrew Wakefield and the measles, mumps, and rubella (MMR) vaccine.  In 1998, Wakefield et al. published an article that implied a link between the MMR vaccine and a ‘new syndrome’ of autism and bowel disease [Wakefield, A. J., Murch, S. H., Anthony, A., Linnell, J., Casson, D. M., Malik, M., Berelowitz, M., Dhillon, A. P., Thomson, M. A., Harvey, P., Valentine, A., Davies, S. E. and Walker-Smith, J. A. (1998). Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. The Lancet 351: 637-641]. In the decade following the paper’s release, subsequent studies found no evidence of a link between the MMR vaccine and autism. Following claims that the data reported in the 1998 paper were falsified, several subsequent investigations proved the data to be false. The paper was retracted in 2010. Despite the retraction, the fear of a potential link between the MMR vaccine and autism had rapidly spread around the world, with parents questioning the risks of vaccinating their children against measles, mumps, and rubella.

For more information on the case of the MMR vaccine research fraud, go to:

  • Deer, B. (2011). Secrets of the MMR scare: How the vaccine crisis was meant to make money. BMJ, 342:c5258 (http://www.bmj.com/content/342/bmj.c5258.extract)
  • Deer, B. (2011). Secrets of the MMR scare: The Lancet’s two days to bury bad news. BMJ, 342:c7001 (http://www.bmj.com/content/342/bmj.c7001.full)

3.5.1 Research misconduct

Section 10 of the Code(http://www.nhmrc.gov.au/publications/synopses/r39syn.htm) lists as examples of research misconduct:

  • Fabrication of results
  • Falsification or misrepresentation of results
  • Plagiarism
  • Misleading ascription of authorship
  • Failure to declare and manage serious conflicts of interest
  • Falsification or misrepresentation to obtain funding
  • Conducting research without ethics approval as required by the National Statement on Ethical Conduct in Research involving Humans and the Australian Code for the Care and Use of Animals for Scientific Purposes
  • Risking the safety of human participants, or the well-being of animals or the environment
  • Deviations from this Code (http://www.nhmrc.gov.au/publications/synopses/r39syn.htm) that occur through gross or persistent negligence
  • Willful concealment or falsification of research misconduct by others.

It is important to note that the examples of research misconduct listed in the Code (http://www.nhmrc.gov.au/publications/synopses/r39syn.htm) are not universally accepted as research misconduct. For example, in the US, the ORI defines research misconduct as “fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results” (http://ori.hhs.gov/misconduct/definition_misconduct.shtml). As discussed in subtopic 1.2, this example again highlights the importance of including statements on research conduct and misconduct in any international agreements.

National and international research codes of conduct stress that misconduct does not include differences of opinion or honest differences in judgement in management of the research project. Honest errors that are minor or unintentional can occur, but, as breaches of the code, they must be acknowledged and rectified. This is particularly so if erroneous data has been disseminated and contributes to the body of knowledge in a field.

Deviation from the Code(http://www.nhmrc.gov.au/publications/synopses/r39syn.htm) is taken to constitute research misconduct, and this module has considered previously the damage to reputations of individuals and the university, together with the undermining of public trust, which can follow research misconduct. Well-reasoned responses to inevitable ethical dilemmas that arise during the course of research can resolve ethical conflict. Such responses not only label issues but identify the underlying elements so that ethical research can be fostered.

3.5.2 Questionable research practices

Increasingly, the question is being posed whether serious misconduct is the only problem which can damage the integrity of the research process. In an early US National Academy of Science publication, Responsible Science Ensuring the Integrity of the Research Process, a distinction was drawn between the three issues of falsification, fabrication, and plagiarism that have underpinned research misconduct definitions and questionable research practices. While these issues and distinctions were initially identified within the context of scientific research, they apply across all research disciplines. Although these practices may not directly damage the integrity of the research process, they can:

  • Erode confidence in the integrity of the research process
  • Violate traditions associated with research
  • Affect research conclusions
  • Waste time and resources
  • Weaken the education of new researchers.

Questionable research practices which are the forerunners of the misconduct defined by the Code (http://www.nhmrc.gov.au/publications/synopses/r39syn.htm) can be found in the planning, conduct, interpretation, publishing, and review phases of research and can include:

  • Bias in selecting methods that favour the desired outcome in results
  • Failing to disclose all conflicts of interest
  • Supplying overoptimistic interpretations to review committees
  • Failing to follow protocols, particularly in human subject research
  • Supervising inadequately
  • Using inappropriate statistical methods
  • Bias in selection of data and controls
  • Drawing unjustified or unsupported conclusions
  • Giving honorary or ghost authorships
  • Withholding crucial information in abstracts or publications
  • Giving cursory review to the work of others
  • Failing to maintain confidentiality
  • Bias towards or against colleagues, fields, or legitimately argued points of view. 

Activity 5 (Safeguarding the Murray–Darling case study: "Correcting an error")

What would you do and why?

You may wish to consider the following set of circumstances and consider how you would choose and defend your point of view. The circumstances are based on an actual case [Nature 417: p677, 2002] but will be put in the context of the Safeguarding the Murray–Darling case study. If you have not already done so, familiarise yourself with it before reading further.

You buy an off-the-shelf statistics package and use it on an unchecked supplier's default setting. The statistics package was used to estimate the health risk posed by particulates in the water in the Basin, which has become a source of fierce environmental controversy. You subsequently read that one research group has recognised the potential for error in using the default setting and have published this as a caution. It is expected that all groups using this methodology throughout the world will revise estimates of the health risk and correct any published work. This is despite the effects it would have in lessening the impact of any emission control research data compiled using the default setting (as in some cases it doubled the estimate of health risk).

1. Do you consider there was any misconduct on the part of your group in using the initial statistical package?

2. What steps will you take to revise your estimates presented recently to the Murray–Darling Management Trust (MDMT)?

Spend no more than 10 minutes on this, make a note of your responses, and bring them to the workshop, where this will be discussed further.

Questions for further reflection

Perhaps you consider that there would be too much fallout in explaining and admitting this error to the MDMT, as you have been working hard to establish a ‘good’ relationship and they have used this data to push for stronger water control measures.

Is this research misconduct?

MDMT are so disturbed by the uncorrected data that you are invited to a board meeting and asked to work on an enviro-friendly campaign and to monitor the outcome of a major remediation effort. You decide it is time to make the correction to the statistics package but use the correction only on the remediation data as it will make the remediation appear more compelling.

Is this research misconduct?

A new Fellow is recruited to your department. He has used this statistical package in the UK and wrote the retraction/reassessment document of his group's work. He wants to share war-stories and asks how you handled the situation and can he see the data from the pre- and post-remediation studies. You withhold the data.

Is this research misconduct?

< Previous Next >