This case discusses basic concerns about confidence in one's own ability to judge the merits of his or her work; the timing of research and proposal writing; honesty; trust between researchers; and trusting the integrity of colleagues in the scientific community.
From: Graduate Research Ethics: Cases and Commentaries - Volume 1, 1997
edited by Brian Schrag
Dr. Susan Ness has been working diligently as a post-doctoral student in material science at State University for the last year. She has been developing a new material, which remains superconductive at higher temperatures than any other measured substance. Ness has almost completely expended her post-doctoral fellowship, but her sponsoring faculty member, Dr. Black, has indicated that he would like to submit a joint research proposal to the National Science Foundation (NSF) for funding to support the costly final conclusive tests on the new material. Ness has not yet published or presented anything on her work, and Black wants to make her results known immediately. Unfortunately, she will not have an opportunity to present preliminary results publicly before the NSF proposal deadline she has targeted with Black.
NSF has indicated that superconductivity research is a top priority, and many people have applied for funding. Ness is aware of the competition yet is confident that they have a very strong proposal. Ness and Black have divided the work; he is overseeing the project, and Ness is documenting the experiments. Ness clearly describes the methods she has developed to create the new superconducting material but then has second thoughts. Knowing that NSF makes use of peer review, Ness realizes that other active superconductivity researchers will read her proposal with much interest and may "steal" her ideas. The implications for her future career would be devastating.
Ness decides to introduce an error into the NSF proposal's otherwise flawless methods description. She believes that the deliberate error will be hard to identify and will prevent replication of the new material. Ness decides not to confide in Black.
Three months later, Ness and Black are awarded a prestigious $250,000 three-year grant from NSF. Black is thrilled and congratulates Ness for her outstanding scientific contributions. He suggests that Ness will have no problem job hunting elsewhere, but he says that he wants to extend an offer for her to stay at State University. The NSF review panel provides feedback to Ness indicating they intend to support her throughout her career and routinely destroys all proposals in their possession. Ness publishes her preliminary results and a clear error-free description of their method in a rapid publication journal and receives scientific acclaim.
Now suppose the case continues:
Ness receives an E-mail from a graduate student at another university who is doing superconductivity research but has been unable to replicate her results. She discovers that the student knows her NSF proposal information, is unaware of Ness's published research and is stymied by the error Ness introduced in the NSF proposal. If Ness references her published research, she feels it will become obvious to the student that she had introduced the error on purpose.
Posted 13 years ago
Michael Pritchard Western Michigan University
This case raises basic concerns about confidence in one's own ability to judge the merits of his or her work; the timing of research and proposal writing; honesty; trust between researchers; and trusting the integrity of colleagues in the scientific community. Before discussing possible changes that might be made in the case, I will elaborate a bit on how these basic concerns arise in the case.
Ness seems to exude confidence in her ability to judge the merits of her own work. By disguising what she takes to be a key element in her process she is, in effect, denying reviewers the opportunity to judge for themselves. At the same time, she seems rather confident that the reviewers will not be able to detect the flaw in the process she describes Thus, by handling the situation as she does, she is resisting a basic premise of good science i.e.., that one should be willing to subject one's work to the critical scrutiny of one's peers. She acts partly from fear, but she also seems to display a kind of arrogance (that she knows and that others cannot even detect her deliberate error). This arrogance is not a good sign, as it is unlikely to be a one-time attitude. Furthermore, she is taking a rather large risk that her reviewers will miss her error. If they do detect it, it is she, not they, who will be seen as lacking; she may be suspected of deliberate deception. She fears that someone might steal her ideas; she distrusts NSF's system of review. Does she have any evidence that NSF reviewers will behave unethically? Does Black? Has she actually talked about her worries with anyone, or is she a loner?
By not consulting with Black, her sponsor, Ness makes two mistakes. First, as a collaborator, he is entitled to know what she is doing. He may not be familiar enough with the processes to find the error, but he is certainly capable of understanding that she intends to deceive the reviewers. She chooses to deceive him as well -- an ethically problematic move and another indication of her distrust of other members of the scientific community (in this case, even a close colleague). By taking this action, she effectively blocks herself from getting advice from Black about how she should proceed. That is the second mistake. Essentially, she is attempting to go it alone, betraying both her collaborator and the scientific community. Of course, even if Black were to approve of her tactic, that would not make it right; but at least she would not have deceived her sponsor. More likely, however, Black may suggest alternatives that are more acceptable ethically.
Here is a possibility: Ness could indicate to the NSF that she is leaving out a crucial part of her process (rather than introducing an error) because she is concerned about others adopting her process before she has been able to demonstrate its reliability. She could offer to provide a fuller account if she is given assurances that reviewers will not use it themselves or reveal it to others. Or perhaps she could tell NSF that she has deliberately introduced an error in order to prevent others from stealing her ideas. If that truly is a good idea, she should be able to convince NSF that it is. However, it is unlikely that NSF would accept either of these alternatives unless it believes that its reviewers are capable of assessing the true merits of the proposal despite the crucial omission or the inclusion of an error. I should think that NSF would reply to either suggestion by saying that it is unwilling to award a $250,000 grant for a project that it is not given the opportunity to evaluate fully. "Trust us or go elsewhere" is the likely NSF response. If Ness were to imagine herself to be in the position of NSF and its reviewers, perhaps she could appreciate this response. Here is the message NSF would hear if it learned of Ness's tactic: "I know what I'm doing. Trust me. Of course, I don't trust you. So, I'm going to lie to you, but only this time. Trust me. I know that peer review and evaluation is essential to good science -- but not now, just as I'm getting started."
Much of Ness's difficulty in this case pivots around the timing of her research. She and Black are worried about meeting a deadline. NSF has deadlines every six months. Ness's best course of action may be to wait another six months. Compromising in order to meet deadlines is a temptation she may frequently face. Her response to pressure in this case is not a good precedent. We might also question Black's rushing Ness ahead.
In answer to the questions for Part A, I would say that Ness has done something unethical. She has deceived both Black and the NSF, without apparent justification. She expects to be trusted to do quality research (presumably without further deception), but she is unwilling or unable to trust those from whom she expects assistance (Black and the NSF). From an ethical perspective, that is a very unhealthy environment. On the basis of what is presented in the case so far, it is not clear whether Ness is justified in doubting the confidentiality of the NSF peer review process, but if researchers cannot trust the NSF process, whom can they trust? After all, NSF is a federal agency supported by our tax dollars, not private industry. Insofar as researchers have such doubts (justifiable or not), it seems that the NSF should do its best to ensure that these doubts are not justified. That is, NSF has an ethical responsibility to ensure confidentiality.
This case illustrates the importance of trust among scientists. What could Ness have done differently to protect her research proposal? Probably no course of action would guarantee the security of her ideas. But she has no guarantee that things will work out well by introducing the error, either. In fact, it seems that her own choice is much riskier. Part B illustrates only one of several possible ways in which matters may go very badly for her.
What occurs in Part B is rather ironic. Ness's ideas are misused (the confidentiality requirement is betrayed by an NSF reviewer), but the error is detected nevertheless. Admitting to the student that she deliberately introduced an error does not end her obligations. She should inform NSF, and perhaps the scientific community. Why should she take this action? There is little reason to think that her proposal has been shared only with this student, who, after all, was not one of the reviewers. Who else has been privy to the proposal? And what sort of mischief might result from others adopting the flawed process? Materials and money will have been wasted. Perhaps there are other risks. It is impossible to know where the misinformation and its consequences will stop. Part B simply gives us one possible variation on the basic theme that Ness has made a big mistake.
I think the case is basically well written. However, it seems a bit exaggerated. It is unlikely that NSF will promise career-long support for a post-graduate (or anyone else). Ness's belief that NSF might respond in this way may be more realistic. Perhaps there is a way of helping readers see that sometimes young researchers exaggerate how much is at stake, just as they may underestimate the importance of extending trust to others. Also, after readers have wrestled with the case, there could be some discussion of the NSF review process (how reviewers are selected, how assurances of confidentiality are provided, and what might happen to reviewers who are caught misusing the proposals they review). That would be a discussion of a rather complicated, but important, web of trust that NSF is counting on.
Finally, some questions arise about the roles of various players in this scenario. Black's role is not entirely clear. If he makes the application to NSF, why does Ness get the award? What are Black's responsibilities? What does providing "administrative information" involve? Are no other researchers involved in this project? If not, and if Black does not understand the science involved, what basis does he have for thinking that Ness is ready to do the research? If he is committed to science (and the responsible use of NSF's money), it seems that he should want the NSF to have as much accurate information as possible about what Ness is doing. It is to be hoped that he would indicate this need to Ness if she asked him about the advisability of deliberately introducing an error. It is also to be hoped that Ness would be able to reach this conclusion independently if she asked herself how he might respond (thus stopping her plan without even having to discuss it with Black).
However, it is possible that Black is too ambitious for Ness (or himself, if some of her success will be credited to him). Ness may be too ambitious. This case provides a good opportunity to discuss the various ways in which personal ambition may tempt researchers to engage in ethically questionable behavior even though they may sincerely think they are justified in doing so (e.g., because they do not believe they can trust their colleagues and peers to act appropriately).
From: Graduate Research Ethics: Cases and Commentaries - Volume 1, 1997 edited by Brian Schrag
Posted 13 years and 1 month ago
At least two levels of analysis are possible here -- legal and ethical. Legally, principal investigators and co-principal investigators must sign the following statement: "I certify to the best of my knowledge that the statements herein (excluding any scientific hypotheses and scientific opinions) are true and complete. . . . I understand that the willful provision of false information or any other communication submitted to NSF is a criminal offense." Thus, despite a possible reaction that Ness's action may be a "small murder," which is bad only in that it will lead to worse actions if it succeeds, the introduction of the error is in itself a legally prosecutable offense.
However, Ness was not caught in this deliberate act. Even if she were caught, it would be difficult to prove her guilt, due to the burden of proving intent and lack of negligence and providing expert witnesses.
That leaves open the question of whether the introduction of the error was ethical. Ness has a responsibility to herself and her co-principal investigator, Black, to be truthful so that her proposal can fairly evaluated. She also has an ethical obligation to other proposers and the NSF peer review process. By being untruthful, Ness has jeopardized her own career and Black's career, penalized the careers of other proposers and corrupted the review process. That may seem like a strong indictment, but the entire system is built upon truth. Violations cannot be taken lightly because they bring into question the entire system.
Does it matter that Ness introduced an error in the method and not a fudging of data? An argument can be made that they differ in severity, but both are misconduct.
Do you think Ness can later claim she made an honest mistake in the method and is thus innocent of misconduct? This possibility will require a convincing explanation that is hard to express unless one is completely truthful.
Mention must also be made about the division of labor in writing the proposal. Although it is common practice to divide writing the parts of the proposal, ultimately all coauthors are responsible for the entire proposal; coauthorship of proposals is similar to coauthorship of a paper. Thus Ness and Black should have ensured that each of them completely understand the other's written contribution; specifically, Black should have understand and caught Ness's deliberate error. Black can be faulted for failing to ask Ness to explain her methods until he understood; Ness can be faulted for not giving an explanation or giving an unclear or misleading explanation.
Back to Top
Ness may or may not have good reasons to support her beliefs, but she has the right to her own opinion. I do see a problem, however, in Ness acting on her belief that the NSF peer review process is corrupt. By introducing an error in the proposal, Ness is actually contributing to corruption of the process. She may be inaccurate in her belief and the NSF peer review process may be of the highest integrity; by her actions, she is creating a self-fulfilling prophecy. If Ness is accurate in her belief, a common response may be that anything is justified in a corrupt process -- anything you can get away with, that is, and Ness has gotten away with an error that has protected her research ideas (or at least that is what she thinks). Ness knows herself what she has done, and her deception will likely manifest itself sometime. What would Ness think if she were a future proposal peer reviewer who had detected such a deliberate error?
I state in the discussion of Question 7 that Ness has an obligation to fix any problems she encounters and not just point the finger while acting unethical herself.
The first thing that jumps out is that this new material or the process to create the new material may be patentable, protected under intellectual property law. There are many ways to apply for a patent (these procedures are changing rapidly), and the best advice is to consult a specialty lawyer. State University will surely have a claim (legitimate or not), and the State University legal department may be one place to start, although the counsel there will be acting on behalf of the University and not on Ness's behalf. Laboratory notebooks and other documentation become very important in patent applications. World patent law is moving toward a global "first-to file" standard (as opposed to the traditional U.S. first-to-invent standard), so time is of the essence.
Patenting the new material or the process to make the new material may conflict with Ness's desire to submit a proposal to NSF and/or publicly present her preliminary results. She may be forced to decide whether it is in her interest to patent or publish; these options may be mutually exclusive.
The best way to avoid a situation in which the researcher feels vulnerable to being scooped is to publish research, to build a body of work that can stand for itself. If it is good work, others will notice and possibly pursue some of the ideas. Researchers should be honored when others follow their work. Most papers are not referenced and are read only by their authors and close colleagues. Problems of proper acknowledgement of intellectual contributions may arise, but science is a self-contained community; most problems will not only be noticed and remedied but the perpetrator's reputation will suffer severely. To a large extent, Ness has been forced into this fear of being scooped because she has not published.
One suggestion is that Ness only sketch her methods and state that the complete methodology is available upon request, but what happens if many people make requests? Ness can also be vague in her description and omit information about her methodology, but that will cause two problems: 1) It may hurt her evaluations if peer reviewers realize that she may not have a completely workable experiment. 2) It also causes ethical concerns if Ness strays close to untruthfulness (e.g., omitting vital information). A last suggestion is that Ness can use specific jargon that is truthful but cryptic and undecipherable. Using specific jargon will probably harm her chances for funding. Unfortunately, many researchers write in jargon not to protect information but because they cannot express themselves clearly in written communications.
The peer review process is handled differently by different organizations. One common practice is to distribute proposals under consideration to researchers at universities and other institutions so that their feedback can be incorporated in the selection process. An explicit statement of confidentiality stating that the prohibits sharing of proposal contents. Unfortunately, professors commonly discuss the proposals with students, possibly even involving students in creating the proposal feedback to be sent back as peer review. After the peer review process concludes, the funding organization notifies all reviewers that the proposals they have in their possession should be destroyed. Unfortunately, many reviewers hold proposals for later reference. Because students who are given proposals by their professors are often unaware of the context of the research material they have been provided, the scenario depicted in this case is not that far-fetched.
Other problems from student involvement in the peer review process include: 1) feedback to the proposer(s) and possibly even award selections may be partially based on student review and not peer review; 2) when students become aware of their covert involvement in the peer review process, they may come to doubt the entire proposal process, perhaps rationalizing unethical actions on the belief that peer review cannot be trusted. On a final note, there should be an assumption of innocence until all the facts become clear. The graduate student in this case appears to be an innocent party, and the person(s) who provided him with the confidential NSF proposal appear to be guilty, but we do not know all the facts.
Ness can either admit her actions to the student or continue to cover-up. her actions. If she continues to try to cover up, the coverup may actually become as significant or even eclipse her original deceit. Of course, any admission will need to be complete, exhaustive and reciprocal, since Ness needs to know how the student came upon her proposal.
Does wrongdoing by another party ever justify wrongdoing by a researcher? In this case, does it really matter whether the NSF peer review process actually leaked confidential information? Specifically, is it introduction of the error unacceptable if the NSF peer review process has integrity and acceptable if the NSF peer review process lacks integrity? Ness must examine her own actions independent of the actions of others. A researcher cannot control the actions of others and should concentrate on her own ethical behavior. To do otherwise brings ethical behavior to the lowest common denominator.
I see at least two other major ethical obligations: