Article Text


Bad behaviour does not equal research fraud
  1. Bob Williamson1
  1. 1The Murdoch Childrens Research Institute, 10th Floor, Royal Children's Hospital, Flemington Road, Parkville Vic 3052, Australia; williamb{at}
    1. D Geggie2
    1. 2Accident and Emergency Department, Arrowe Park Hospital, Arrowe Park Road, Upton, Wirral, Merseyside CH49 5PE, UK; DAVID{at}

      Statistics from

      I was not impressed by Dr Geggie's article offering a survey of the attitudes of newly appointed consultants towards research fraud (Journal of Medical Ethics 2001;27:344–6). Indeed, by mixing up categories of misconduct from what is at most “bad behaviour” to the very serious, he is not entirely beyond reproach himself. I remind readers that Dr Geggie suggested that 55.7% of the respondents had observed (from the title) “research fraud”.

      If the term “research fraud” is to have any meaning, it must be reserved for conduct that consciously and deliberately attempts to impose a fraud on others. The US National Academy of Sciences' report, On Being a Scientist: Responsible Conduct in Research distinguishes clearly between “misallocation of credit, honest errors, and errors caused through negligence” and “deception, making up data or results, changing or misreporting data or results, and plagiarism”. The former are “ethical transgressions . . . that generally remain internal to the scientific community . . . dealt with locally through peer review, administrative action, and the system of appointments and evaluations”. The latter “strike at the heart of the values on which science is based”. The White House's Office of Science and Technology Policy reached similar conclusions, restricting research misconduct to “fabrication, falsification and plagarism”. I agree with these assessments.

      In Dr Geggie's paper, deception would include deliberate falsification of data (category 3 of Dr Geggie's table 1), cheating (4A) and deliberately plagiarising without attribution (4D). We then move into fairyland, because Dr Geggie next asks “have you ever been an author on a paper for which any of the authors have not made a sufficient contribution to warrant credit for the work” or “has your name ever been omitted from a paper for which you had made a substantial contribution”, and, surprise, surprise, over half answered yes. These are examples of, at worst, ethical transgressions (but more often examples of delusions of grandeur on the part of the offended) but are not research fraud.

      Those of us who run large research groups know from experience that authorship of any paper poses problems. I have never met anyone, from the cleaner to the director, who undervalued his/her own contribution to the work! The issue arises with such consistency that many labs (including my own) have long documents on how to resolve the problem of who goes on, and who stays off, the authors' list. But research fraud? Give me a break!

      Of those who responded, approximately 10% had personal knowledge of research fraud, and 5% at most (it is not possible to tell if there is overlap between those answering different questions) admitted to acts that could be regarded as research fraud. This corresponds to my own experience. I have spent more than 40 years in human molecular genetics, a very competitive field, of which 25 years has been spent running very large groups. I know the work of over 1000 scientists well. I have personal knowledge of one case of serious research fraud and perhaps three or four cases where the reporting of data was so careless as to verge on misconduct (though without intent to commit fraud). I also know of one case of plagiarism. To blur the boundaries between “research fraud”, “research misconduct”, “possible future misconduct”, and complaining about not getting your name on a paper does no one any good, and certainly does not help any rational discussion about research ethics.

      Author's reply

      Professor Williamson makes a valid point about the term “research fraud” and I agree that the term covers a number of different categories of unethical behaviour. I also pointed out that “Research fraud can take many forms” in the discussion section of my paper (Journal of Medical Ethics 2001;27:344–6). For the purposes of my article I stated that consultants who had answered “Yes” to questions 1, 2, or 3 of table 1 had reported “observed misconduct” and I stand by this conclusion. The dictionary on my bookshelf (Concise Oxford Dictionary (6th ed)) defines “misconduct” as “improper conduct, esp. adultery; bad management” and I think that what the consultants were reporting in the questionnaire fits this definition. The same dictionary includes in its definition of “fraud” the phrases “use of false representations to gain unjust advantage” and “person or thing not fulfilling expectation or description”. The behaviour described in questions 1, 2, and 3—that is inappropriate authorship, omission of names of contributors who made a substantial contribution, and the intentional alteration of data for publication, fits this definition. In the article I tried to use the term “fraud” as a more general term than “misconduct” and I apologise if this caused confusion. I think, however, that the majority of readers (and certainly the majority of my colleagues) would have had something approaching the above definitions in mind when they read the article (although several colleagues have told me to get a more up-to-date dictionary!) Whilst I acknowledge that Professor Williamson makes a valid point, I do think there is a danger of any rational discussion becoming bogged down in semantics and I think it is clear from my article what the initial data was and how I have derived my variables.

      Professor Williamson dismisses the medical consultants (not cleaners!) who answered yes to questions 1 and 2 as having “more often delusions of grandeur” on their part. That is a point of view—but I suspect the majority of respondents to my questionnaire do not share it. Indeed I think his attitude risks sweeping a very real problem under the carpet and, whilst I am aware of attempts to standardise authorship, I think this is a problem which requires further investigation.

      I find the overall tone of this letter and phrases such as “into fairyland” and “give me a break” inflammatory and this, unfortunately, distracts from some valid points.

      Professor Williamson's main criticism appears to be that he has learnt nothing new from my paper. The original idea for this study came after a lively discussion in a bar about the prevalence of research fraud (a subject of which I knew very little). After this discussion I did a literature search and could find only a few references to research fraud and misconduct. I decided to perform the study in an attempt to answer the questions I had. I wish that Professor Williamson had been present at the original discussion—he could have saved me a lot of trouble! On a more serious note, I am sure that the professor is aware that in the “hierarchy of evidence” questionnaire surveys rank higher than anecdotal evidence and personal experience (although only just!). The value of my article is that it adds evidence to a subject in which there is a marked paucity of hard facts.

      I am sorry that Professor Williamson was unimpressed with my paper. Unfortunately, the very nature of the subject of research fraud/misconduct mitigates against good quality research. I am sure that readers would welcome any double-blinded randomised control trials (RCTs) on the subject and, if the professor wishes to carry out some research in this area I am sure that the scientific community would welcome it.

      View Abstract

      Request permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.