Scientific research is a process that requires scientists to perform, interpret, and communicate results in an unbiased manner, which excludes any thoughts of self-interest. This is the researcher’s responsibility to their peers and society at large, which is rewarded by trust and an atmosphere of freedom in research with little interfering regulation. However, cases of research fraud undermine this trust, and may lead to the institution of overbearing policies that restrict the freedom researchers have enjoyed. This scenario may be closer to reality than one would think due to a number of relatively recent and highly publicized cases of research fraud.
Research fraud is defined, by the Office of Science Technology, as the fabrication, falsification, or plagiarism in proposing, performing, reviewing research, or in reporting research results. An example of fraud that matches this definition perfectly can be found in the case of Jon Hendrik Schon, a Bell labs physicist. Schon was an upcoming physics prodigy who was already being labeled as a future Nobel Prize winner due to amazing contributions to semiconductor research and nanotechnology. His contribution included ~90 publications in a span of two years (averaging one every eight days), often in Science or Nature. However, the keen eye of one Cornell physicist quickly turned events around when she noticed that in two separate publications Schon had used the same figure to represent different results. This observation led her to perform a more thorough search of his published papers, which resulted in the discovery of other instances of the same type of misconduct. This prompted an investigation of Schon, which found that there was enough evidence to conclude that he had forged results in 17 of 25 papers investigated. This resulted in Schon being fired from his job at Bell labs, and in conjunction with the other recent cases of fraud (1, 2) has prompted the scientific community to begin asking some hard questions about dealing with fraud.
One question raised due to these cases of fraud surrounding Schon and the others was who shares responsibility with them? Should the direct supervisors of these scientists be held responsible? When such sensational results are reported, is it not the supervisor’s responsibility to demand extraordinary validation knowing the likely scrutiny these results will see? Should co-authors be held responsible for all the data presented in a publication? Can we expect co-authors to be experts in diverse fields that may come together to publish a paper and make them responsible for validating all the data? Many argue that collaboration is built around a foundation of trust and if we take such measures than collaborations may no longer work. Finally, in the case of Schon’s publications there were questions raised about the responsibility of the journals involved. Specifically, there were allegations that a few of Schon’s most extraordinary publications were given special treatment by the journal editors, and that some papers may have bypassed the review process completely so that these journals could publish these incredible results more quickly.
The second major question to deal with was what to do with an author’s publications in cases of fraud like that of Schon’s. Should all of his publications be retracted? Should a committee be put together to investigate and validate all these publications? Should they simply be left to come under review by other scientists and allow future publications to support or reject these papers? In truth, none of these options are perfect because not all of Schon’s work or that of his co-authors may have been fraudulent; therefore, simply removing all his work may be unjust. As for forming a committee, a committee requires money and time of scientists, both of which is often precious. Finally, review by the scientific community often takes a very long time during which these publications could influence research of unknowing scientists wasting both time and money. In Schon’s case, individual journals have retracted some of the publications or appended the publication with the investigating committees report on Schon’s fraudulent activities.
These and the many other cases of fraud are not the first and are not the last to confront the scientific community. The community has done its best to deal with misconduct but it is still not proficient at dealing with cases of fraud. However, this started to change with the creation of the Office of Research Integrity (ORI) in the US. ORI was established as an independent entity in 1993 and is charged with the job of developing policies, procedures, and regulations related to the detection, investigation, and prevention of research misconduct. ORI reviews and monitors research misconduct investigations conducted by applicant and awardee institutions, and recommends research misconduct findings and administrative actions. ORI is also responsible for implementing activities and programs to teach the responsible conduct of research, promote research integrity, prevent research misconduct, and improve the handling of allegations of research misconduct. The establishment of such an entity is very important to future research due to the estimated prevalence of misconduct that is summarized below.
Documented cases (1 in 10,000)
Know of an undisclosed case (1-13 in 100)
Major deviation found in audit (1 in 10)
Misrepresentations in fellowship applications (1 in 5)
Students willing to fake data (1 in 2)
In essence, this survey suggests that research fraud may be much more common than reported, and this will undoubtedly lead to more cases becoming public resulting in a potential erosion of public trust. For this reason, the ORI and the establishment of similar groups in other countries will be very important in showing the public that the scientific community strongly disapproves of misconduct, and will deal promptly and severely with it. The reason this is so important is that the public pays for most public research through their taxes, and therefore have influence over the amount of funding given to specific areas of research and research agencies through their votes – setting up the possible situation where the lost of trust in scientific research could be recapitulated in the loss of research dollars. Moreover, basic research represents the foundation of future treatments, therapies, and drugs that will be a used by the public for treatment of disease. For the public to accept these treatments they must have trust in where they came from.
1. Taubes, G. (2002). The strange case of Chimeraplasty. Science. 298,2116-2120.
2. Dalton, R. (2002). Misconduct: the stars who fell to earth. Nature 420(6917): 728-9.