CLOSEOUT FOR M94090030 ' - - subject's2NSF proposal3 for merit review. OIG's investigation report and NSF's Deputy Director's 30 September 1997 letter to the subject reflecting his decision constitute the closeout for this case. cc: AIG-Oversight, IG I .. - The corn~lainantid in the Department ' The subject's NSF propo _ declined. ' Page 1 of 1 M94-30 NATIONAL SCIENCE FOUNDATION 4201 WILSON BOULEVARD ARLINGTON, VIRGINIA 22230 April 9 , -1998 CERTIFIED MAIL-RETURN RECEIPT REOUESTED - Dear - On February 25, 1998, NSF received your appeal of the two-year debarment which the Acting Deputy Director of the National (NSF) imposed upon your client, ased upon his misconduct in science. In your appeal, you acknowledge that misconduct in science when he plagiarize .proposals, as well as two National Institute of Health (NIH) grant proposals that he submitted to NIH. You are, however, I committed in two NSF grant appealing the debarment sanction imposed by NSF, contending that it is too harsh a sanction for your clientst misconduct. In your view, your client's conduct constituted an isolated event rather than a pattern of miaconduct because all of his actions were done in the pursuit of funding for one project. In support of your argument; you cite to the University (the nUniversityn) Investigative Scientific is conduct Report in which the Committee concluded that your clientrs plagiarism was single incident in his career.' As we noted in the Notice of Proposed Debarment and the Debarment Notice, your client sequentially submitted four distinct proposals, albeit for the same underlying research project. Your client plagiarized from two different source documents and each of the four proposals contains different plagiarized text. In . , N S F t sview, this does constitute a patterh of misconduct. We also note that the University's Investigative Committee was only aware of the 1994 episodes of plagiarism when it issued its report. During its investigation, the Committee asked your client if he had ever plagiarized on other occasions and your client erroneously stated that he had never done so on before. After the University had completed its investigation, however, NSF1s OIG discovered that your client has also plagiarized text :: from a review paper authored b - y in his 1993 NIH and 8 ,. :NSF proposals. When the Committee characterized your client's conduct as a single incident, it was unaware of your client's additional plagiarism in the 1993 NIH and NSF proposals. As we previously noted, the administrative record does not 6 to I I have considered the arguments raised in your appeal and conclude that a two year debarment is warranted. The administrative record establishes that your client plagiarized in .four different proposals, that he plagiarized not only from published material but also from an NSF confidential proposal, and that he attempted to conceal his plagiarism by asking NSF not to use the -original author as a reviewer of his proposal. I also n6ke that the University's Investigative Committee, which was ';?%iaware of the full extent of your client's plagiarism, recommended that your client be barred from serving as a Principal Investigator on a Federal grant or contract for three years (Exhibit 7 to the OIG Report). NSF1s two-year debarment is . shorter in duration than the sanction recommended by the University. Your client is debarred until September 30, 1999. His debarment shall be effective throughout the Executive Branch of the Federal Government. Your client will be excluded from receiving Federal financial and non-financial assistance and benefits under nonprocurement programs and activities unless an agency head or an authorized designee makes a determination to grant an exception. Nonprocurement transactions include grants, cooperative agreements, scholarships, fellowships, contracts of assi~tance,loans, loan guarantees, subsidies, insurance, payments for specified use, and donation agreements. In addition, your client is.prohibited from receiving any Federal contracts or approved subcontracts under the Federal Acquisition Regulations (FAR) at 4 8 CFR Subpart.9.4 for the period of this . , debarment. 45 CFR 620 .I10 (c). Sincerely, - &&- - Neal Lane Director NATIONAL SCIENCE FOUNDATION 4201 WILSON BOULEVARD . 1 ARLINGTON, VIRGINIA 22230 : September '30, 1997 OFFICE OF THE DEPVTY DIRECTOR CERTIFIED MAIL-RETURN RECEIPT REOUESTED Re: Dear - Notice of Misconduct in Science ~eterminationand Proposed - - - f o Debarment This letter and the attached investigative report serve as formal notice that the National Science Foundation (NSF) proposes to debar your client, 4- from directly or indirectly obtaining the benefits of Federal research grants for a period of two years. A person who is debarred will be excluded during the period of debarment from Federal financial and non- - *. financial assistance and benefits under non-procurement Federal programs and activities. See 45 CFR S620.110, 8620.200. In addition, your client will also be prohibited from receiving any Federal contracts or approved subcontracts under the Federal Acquisition Regulations (FAR) at 48 CFR Subpart 9.4 for the period of this debarment. - '45 CFR §620.110(c). Debarment of an individual is effective throughout the executive branch of the Federal Government. Reasons for Debarment NSF's decision to propose debarment is based upon a referral from our Office of Inspector General (OIG).' The Foundation's administrative record indicates that - Jengaged in plagiarism in two proposals that he sObmitted to NSF. in the report .. In addition, the 1994 ubmitted by Department had received as an NSF January, 1994. The faculty member asked portions of the proposal and nature of the proposal. Under NSF1s regulations, "misconduct" is defined to include "plagiarism, or other serious deviation from accepted practices in proposing, carrying out or reporting results from activities funded by NSF. " 45 CFR S689.1 (a). J By submitting proposals to NSF that copy the work of others without attribution, your client misrepresented as his own portions of the proposal pertaining to significance of the proposed research and experimental methods. This affects the integrity of his proposals. Clearly, originality is a critical factor in evaluating a grant proposal. These portions of the proposals were important elements in NSFfs evaluation and review of the prqposals your client submitted. \ -L Your client acknowledges that he copied text from the 0 proposal and published article without permission or proper attribution. In defense of his actions, however, your client stated that "it was not his intention to steal innovative ideas, methods or conceptsM and that he trying to achieve the same clarity of presentation found in the -proposal to express ideas that they held in common. Plagiarism involves using the ideas or words of another person without attrubution. The copied text should have been marked off by indentation or quotation marks and a citation to the original work should have accompanied the text. The fact that your client could not improve upon the clarity of language used in- proposal does not justify plagiarizing his work. In v t he Investigation Committee found that the plagiarized text included a statistical package that was not publicly available. We find that your client's actions constitute plagiarism as well as a serious deviation from accepted practices within scientific community. Accordingly, we conclude that your client committed misconduct in science. Requlatorv Basis for Debarment 1 In deciding what response is appropriate when misconduct is I found, NSF must consider the seriousness of the misconduct; whether it was deliberate or careless; whetherlit was an isolated event or part of a pattern; and whether the misconduct'affects only certain funding requests or has implications. for any application for funding involving the. subject.of the misconduct finding. . 45 CFR 5689.2(b). Severe misconduct is a cause for debarment because (i) it affects the integrity of NSF research or education programs, see 45 CFR 8620.305(b); 45 CFR 5689.l(e), and (ii) it is a "cause of so serious or compelling a nature that it affects the present responsibility of a person11,45 CFR 5620.305 (d). J i - 7 Debarment must be for a period commensurate with the seriousness of the cause. 4 5 CFR 5620.320(a). The burden of proof is on the governmenk to establish facts which justify debarment by a preponderance of the evidence .. :45 CFR 5689.2(dl , 5620,314(c). . According to the ~nvesti~ative ~ e ~ o r tyour , client'-knowingly committed plagiarism when he copied text from the published article into his 1993 and 1994 proposals, and when he copied text n - proposal into his 1994 grant proposal. With from theto t e 1994 proposal, the record indicates that your respect client used the plagiarized materiax from the proposal to address deficiencies-which the peer reviewers identified in his 1993 proposal. Specifically, the peer reviewers criticized the absence of discussion on significance of proposed research and your clientls methodology descriptions in his 1993 proposal. Your client then plagiarized text from the proposal to -% strengthen these portions of his revised 1994 proposal. I Furthermore, in your client's cover letter to his 1994 proposal, he specifically requested that not be used as a reviewer of his proposal for ostensibly conflict of interest reasons. However, this was not the usual practice at Your client's institution. Rather, it is evident that your client made this request because he did not want- to discover that he had plagiarized his proposal. ~ l of l this evidence indicates that youk client knowingly committed plagiarism. In addition, your client's behavior is more egregious because he plagiarized text from an NSF proposal which another researcher in his Department had received in confidence as an NSF peer . reviewer. Although your client was not the designated NSF peer reviewer, he was advised by his colleague of the confidential nature of the proposal. . . Finally, your client's behavi0r.i~also more serious because he engaged in a pattern of plagiarism by submitting four proposals to federal agencies which contain plagiarized text. In addition to submitting two proposals to NSF on separate occasions which contained plagiarized text, he also used much of the same plagiarized materials in two proposals which he submitted to NIH. In your response to the OIG investigative report, you contend that your client's actions are an isolated event rather than a pattern of misconduct because all of his actions were done in pursuit of funding for one specific project. However, your client's submission of four distinct proposals containing. different plagiarized text, albeit for the same underlying research project, do constitute a pattern of misconduct. Accordingly, we are proposing the following remedies: debarred for two years o y w i l l be excluded from participating as an NSF pane ist, reviewer, advisor or consultant for three years Procedures Governins Proposed Debarment/Scientific Misconduct Allesations Under our regulations, your client has 30 days after receipt of this notice to submit - - in person, in writing, or through a representative - - information and argument in opposition to the proposed debarment. 45 CFR §620.313(a). During this 30-day period he may also review the attached ~nvestigativeReport and submit comments or rebuttal. 45 CFR S689.8 (c)(11, 5689.1 (el'. Comments submitted within the 30-day period will receive full consideration and may lead to revision or withdrawal of the Investigation Report or of the recommended disposition. Any response should be addressed to Lawrence Rudolph, General Counsel, National Science Foundation, 4201 Wilson Boulevard, Room 1265, Arlington, VA 22230. If you have any questions, please .-- contact Mr. Rudolph at (703) 306-1060. We are attaching a copy of the Foundation's regulations on Non-Procurement ~ebarmentand Misconduct in Science and Engineering. '/ Sincerely, A C/ Joseph Bordogna Acting Deputy Director Attachments (4) Investigative Report Nonprocurement Debarment Regulations FAR Regulations Misconduct in Science Regulations REPORT OF INVESTIGATION INTO ALLEGATIONS OF MISCONDUCT IN SCIENCE The Office of Inspector General (01G) has concluded &- qh t (the subject) of-eht (the University) plagiarized text and references from a published review article into two proposals submitted to the Federal government, one of which was sent to NSF and the other to the National Institutes of Health. Much of the same material was again plagiarized into two revised proposals he submitted the next year to the same federal agencies. The latter set of proposals also contained text and references he had . plagiarized from an NSF proposal he had received in confidence. OIG concluded that the subject's actions constituted plagiarism and violation of confidential peer review. These conclusions are based on an investigation performed by the University and expanded by OIG. OIG recommends that NSF find that the subject committed misconduct in science and take the following actions as a final disposition in this case. The subject should receive a letter of reprimand from the NSF Deputy Director informing him that he was found to have committed misconduct in science. NSF should debar the subject from receiving federal government grants for 2 years. -.* For the 2-year period following the debarment, when proposals are submitted by the subject or on his behalf to NSF, he should be required to submit certifications to OIG that, to the best of his knowledge, they contain nothing that violates NSF's Misconduct in Science and Engineering regulation. Further, he is required to ensure that his department chairperson, or equivalent, submits an assurance to OIG that, to the best of that person's knowledge, the submitted proposal does not contain any plagiarized material and all source documents are properly cited. He should be prohibited from partlclpating as an NSF reviewer, advisor, or consultant for a period of 3 years from the final disposition of this case. Page 1 in the Department - from the PI's proposal during his confidential review of that document. After we had initiated our inquiry we were informed that the subject's pending proposal J-4R O l at the National Institutes of Health 0 had- been recommended for an award. It was-alleged that this proposal3 also contained text plagiarized from the PI's proposal. OIGVsINQUIRY Proposal Submission Chronology Early in our inquiry we learned that both the subject's and the PI's 1994 NSF proposals were resubmissions of earlier declined submissions to NSF. After our initial communications with the subject we learned that he had submitted proposals to NTH that were similar to those - b he had submitted to NSF. The table below describes the submission chronology for the relevant NSF and NIH submissions by the subject and complainant. ~ - ' The proposal was entitled " It named &ro-eht requested or 3 y m of researc? support. It was declined. * The proposal was entitled, " ." It named the PI as the sole researcher. The PI requested in researcll support for 3 years. On the basis of this proposal NSF Page 2 ' Date of Submission* November 16,1992 September 10, 1958 September 27, 1993 December 20, 1993 . Subject's Submissions ,$ B N ' ~ " 1993 NSF proposal" " 1993 NIH proposal" d Outcome declined declined declined7 funded - 0 PI's Submissions " 1993 NSF proposal" * 6 I 4. ~.iov 4 " 1994 NSF proposal" June 29, 1994 hnded D " 199&NlH proposal" July 14,1994 ' ' B N ~ declined 1 " 1994 NSF proposal" 1,*signed by PI or authorized organizational representative / During our inquiry we compared the subject's four proposals with the PI's and concluded that the subject's 1994 submissions contained text that appeared to have been copied from the 1994 NSF proposal submitted by the PI. Evidence we developed later in this case caused us to change that conclusion. However, in this case, we present the evidence in the order it was evaluated: first by our inquiry, then by the institution's investigating committee, and finally during our subsequent investigation. . L Copying into the Subject's 1994 NSF Proposal On the basis of the allegation of plagiarism we compared the subject's 1994 NSF proposal with the PI's 1994 NSF proposal. We found five sections of text in the subject's NSF proposal that appeared to be identical or substantially similar to text in the PI's proposal. The material was not offset from other text and was not accompanied by a citation to the source document. The five sections also contained six bibliographic citations that appeared to have been copied from the PI's proposal. The sections varied in length from 1 sentence to as many as 22. Sections 1 and 2 appeared in the "Project Summary." Section 3, the proposal's "Prospective Significance," appeared to have been copied in its entirety from the PI's proposal. Section 4 was composed of the lengthy description of an experimental method. The fifth This earlier proposal 0named the PI,% the ~ l investigator4 e It requested to support a 3-.,, ygar researcb p o r t . The earlier proposal the same title as the 1994 proposal. It named the subject as the sole PI and reauested to suuwrt a 3-vear research effort. 7 subject as u. sole PIE The declined proposal requested for 5 years of research support. 1 I Unlike NSF pfoposals, NIH $roposals that e i v e priority #ores thatkliminate them from fuqiing are not declined. They are administratively withdrawn when the revised proposal is submitted. In this report, for I simplicity, we refer b t h e u-ded NIH a@ declined NSF proposals as "declined." 4 , 1 4P 9% i Page 3 section was a two sentence descr~pt~on of stat~sttcal tests. We noted that the subject accompanted h ~ proposal s with a cover letter that included a list of suggested reviewers. In it he 'stated, "Because of confl~ctof interest, I kindly request that you not forward my grant proposal to either the Department o-f at u n i v e r s i t y [a major east coast University] or to [the PI at his institution] . . . ." . . Of the 92 references in the subject's proposal, 2 of them are to papers by the PI In ' contrast, three of the six ad hoc reviewers of the subject's proposal commented on the contributions of the PI to the subject's research field and the absence of citations to the PI's work. The three prov~dedexact citations to the PI's publications in the~rreviews. One said, "The presence of [particular] products in t h e l l l l l ) i s known (as stated in the proposal and in other unreferenced literature e.g [a citation to a publication by the PI])." Another said, - . "This grant also fails to reference the work of [the PI et al.] who have, indeed, studied the presence of [particular]b-& both normal and abnormal [a citation to the PI'S paper is included]." This reviewer attached a copy of the first page of the Prs papers to his * review. The third said, "It is generally appreciated (see [a reference to a review article8 by the PI]) that c e r t a i n r e s p o n d to varoius-. . ." Breach of Confidential Merit Review of the PI's 1994 NSF Pro~osal he PI's 1994 proposal was subrnitted'io NSF in December 1993 and sent out for mail review in early 1994. Among the reviewers was (the Researcher) a facuity member in the same department as the subject. NSF's records show that since 1992 the Researcher had received four proposals subm~ttedby the PI for merit review; he provided reviews on three of these. NSF received the Researcher's comments on the PI's proposal in March 1994. He did not include names of other suggested reviewers or comment that he had shared the proposal with anyone else and the subject has never served as a reviewer for NSF. I since the subject's 1994 proposal contained passages of text that were identical to those in the PI's proposal, it appeared that the confidentiality of NSF's merit review process might have been breached by the subject gaining access to the PI's confidential proposal with or without the Researcher's knowledge when NSF sent the proposal to the Researcher for review. Subiect's 1993 NSF Proposal We found that the subject's 1993 proposal to NSF was an earlier version of his 1994 proposal and was declined. The 1993 panel summary statement said AS discussed later in this report, we learned that the subject had copied text without attribution from this review article into his 1993 sub~nissionsand this text reappeared in his 1994 subtnissions. Page 4 This is potentially an interesting area of research. Unfortunately, the -, investigator did not do a satisfactory job in presenting the background, explaining the significance of the proposed studies and presenting a clear J rationale for each of the proposed experiments. In addition, the details of the . planned experiments are not clearly presented. The Panel recommended that the investigator resubmit the proposal including these and the suggestions by the reviewers. Section 3 of the subject's 1994 proposal, one of the larger blocks of copied text, was directly responsive to the reviewers' comments about explaining the significance of the proposed . research: - ' . The subject submitted a letter suggesting names of potential reviewers for his 1993 proposal. He stated, "Because of conflict of interest, I kindly request that you not forward my i grant proposal to the Department of at [the major east coast University]." The cover letter that accompanied his 1994 proposal (containing the text we found to be copied . from the PI's proposal) included this same request and additionally asked that the PI be excluded as a reviewer. Subiect's Response to OIG's Inquiw about Text Copied into his 1994 NSF Proposal We concluded that the subject appeared to have copied text from the PI's 1994 NSF proposal and that some of the text copied was directly responsive to the reviewers' comments. It appeared as if the subject had modified his cover letter from the previous year to specifically exclude the PI, the original author of the copied text, as a reviewer of his 1994 NSF proposal. We sent the subject a letter requesting information about the allegations of plagiarism and violation of confidential peer reyiew. i In March 1995, the subject'responded to our inquiry letter (see Appendix 1). He said * that "members" in the department had been asked to review the PI's proposal. He admitted that he had copied the five sections of text from the PI's proposal and "acknowledge[d his] error and the allegation of plagiarism in the 5 sections. . . by my use of [the PI's] grant without his explicit permission." He said "it was never p i s ] intention to steal innovative ideas, methods or hypotheses . . . ." He said the PI's and "our ideas overlapped significantly. His style of writing and expression of ideas in some areas were succinct, clear and concise. It was only my desire to achieve similar clarity of presentation that lead [sic] me to make an egregious error, plagiarizing his work." He said, "It was my misguided belief that I was not stealing any new ideas or cutting edge of technology. I admit this belief lead to totally improper conduct." He said if his proposal had not been declined he would withdraw it. He stated, "I can assure the National Science Foundation that this error or a similar error shall never be repeated by me and that all future work will be exclusively my own." Page 5 We asked the subject who wlthln'the department had provided hlm wlth the PI'S proposal. He identified the ~esearch'erand said, "because I was working in a related area of research, I was asked to read several parts of the grant about which I have significant expertise. I had no instructions about the use of the grant other than to review it." He said, "CThe PI'S] grant was exceptionally well written . [I]t was not my intention to steal his ideas, methodologies, or innovative concepts I did use parts of [the PI'S] grant without the knowledge or permission of [the Researcher, and the PI]. . . . Again, the sole purpose was to express in a clear, concise manner ideas we held in common. . . . I can assure you without any reservation whatsoever that such errant behavior will never again be rebeated by me." I We noted that the Researcher was listed, along with four other members of the department, as a "collaborator" on the Subject's 1994 NSF proposal budget page.9 A Current and Pending Support statement and a letter of support from each collaborator were included with the proposal. We asked the Researcher why the subject had been given the confidential proposal and what the Researchei's role was in preparing the subject's proposal. In his written response (Appendtx 2) the Researcher told us that he had given the subject the PI's proposal to review because the proposal contained a descript~onof a technique that the subject "was performing in his laboratory." He said, "I gave [the subject] the original proposal for a few hours and stressed the confidential nature of the proposal. Unfortunately, I had neglected to remember the requirement to notify the lVSF program officer for requesting a colleague's assistance in evaluating a proposal. I had no knowledge of any other use [the subject] had of the proposal, outside of evaluating the . . . technique." The Researcher also said, "I had no part in the preparation of the [subject's] proposal norhid I read any version of the proposal . . . [The subject] apparently included my name as an unpaid collaborator in his budget page and obtained information for the Current and Pending Support Statement pertaining to me from my secretary, unbeknownst to me. I reiterate that I had not then nor since ever received a copy of [the subject's] NSF proposal." The Researcher's comments suggested that the subject may have misrepresented the participation of his colleagues in hls proposed research. Subiect's 1994 NIH Proposal Shortly after we corresponded with the subject about the apparently copied material in his 1994 NSF proposal, we were informed that the subject had received funding for an NIH proposal that appeared, by title, to be in the same research area as his NSF proposal. The PI's 1994 NSF proposal was sent to the Researcher at the subject's institution in January 1994, 6 months before the subject's NIH proposal was submitted and 6.5 months before the subject's NSF proposal was submitted. t We compared the subject's 1994 NIH proposal with the PI'S 1994 NSF proposal and found two sections of text in the subject's proposal that appeared identical or substantially Four of the five collaborators, including the Researcher. were listed as unpaid collaborators. Page 6 similar to text in the PI's proposal. The two sections of text also contained seven bibliographic citations found in the text in the PI's proposal and that appeared to have been copied from it along with the text. The first section, the "Prospective Significance" df the proposal, appears to have been entirely copied from the PI's proposal. It is 23 lines long. The second section is a 17-line description of an experimental method. I We then compared the subject's 1994 NIH and NSF submissions. The copied material in sections 1, 2, and 5 of the NSF proposal is only found in that proposal. The copied material in sections 3 and 4 is found in both the NIH and NSF proposals. However, for each of these latter sections, the subject's NIH proposal contained a few additional copied lines. It appeared that in preparing the NSF proposal the subject had eliminated selected lines to comply with NSF's more restrictive page limit.1° Of the seven copied citations in the NIH proposal, six of them are also found in the NSF proposal. For comparison purposes we have included copies of the PI's 1994 proposal (Appendix 3), and the subject's 1994 NIH and NSF proposal (Appendices 4 and 5, respectively). We have highlighted the 5 sections of copied text and references. The yellow highlighted material is common to all three. The pink highlighted material was copied into just the N w proposal and the green highlighted material was copied into just the NSF proposal. w e have bracketed and sequentially numbered the sections in the PI's proposal and numbered the sections in the subject's NIH and NSF proposals accordingly. We have also asterisked and numbered the seven references in the bibliography of the PI's proposal that appear to be copied and correspondingly numbered the copied references in the subject's NSF and NIH proposals. Subiect's 1993 NIH Proposal \ When we learned that this case also involved copying into an NIH proposal, we. informed the Office of Research Integrity (ON) at the,Public Health Service. Through that Ofice we learned that the subject's funded NIH proposal was a revision of an earlier proposal submitted to NIH in 1993, before the PI's funded 1994 NSF proposal was submitted. We-- reviewed the subject's earlier NIH proposal and the reviewers' comments. Interestingly, like the NSF reviewers of the subject's 1993 proposal, the NIH reviewers said that his earlier NIH proposal failed to explain the expected outcome of the various experiments and their potential significance. They said The rationale for individual experiments is not clearly defined and the focus seems largely centered around detailed methods. Potential pitfalls and alternative approaches are not offered. The application would be greatly strengthened by,a more systematic design in which the experiments to be done 'ONSF places a 15-page limit on research descriptions; PHs places a 25-page limit on the Research Plan description. Page 7 are outlined followed by some prediction of outcome(s), how data will be analyzed and future studies planned. The Application for a Public Health Service Grant (PHs 398) contains instructions for PIS who are submitting a revised application. i. The [revisedJ'application must include an introduction of not more than three pages that summarizes the substantial additions, deletions, and changes. The introduction must also include responses to the criticisms and issues raised in the summary statement. The changes in the Research Plan must be clearly marked by appropriate bracketing, indenting, or change in typography, unless the changes are so extensive as to include most of the text. In the three-page introduction to his 1994 NIH proposal (containing the copied text) the subject said, "I have listed below the changes that have been made in this application to accommodate the criticisms of the reviewers. All new material in the text of the grant has been italicized." In response to the reviewer criticisms he said he had rewritten many of the experimental methodologies. Those rewrites, including the text we identified as section 4, were italicized. In his introduction he commented that he had added a section entitled, "Prospective Significance." This section, copied from the PI'S NSF proposal (our section 3), was also printed in italicized text. Conclusion of OIG's Inquiry ! After reviewing all of the evidence and the subject's and Researcher's responses to our inquiry letters, we concluded that an investigation was warranted: there appeared to be sufficient substance to the allegations of plagiarism, violation of the integrity of NSF's confidential merit review process, and misrepresentation. OR1 concurred, but viewed its case as limited to the allegation that text in the subject's 1994 NIH proposal had been plagiarized. We agreed to separately defer our cases and that these deferrals could be addressed by the University in one investigation report. There was no need for separate reports. Consistent with NSF's position that awardee institutions bear primary responsibility for preventing and detecting misconduct we informed the University of the allegations and, at its request, we agreed to delay any possible investigation by our office until the University had completed its own, independent investigation. We formally referred an investigation into this case to the University in an August 1995 letter. Page 8 Although not required by NSF's regulation, the University's misconduct policy"1 requires that it conduct an inquiry before it initiates an investigation. The University's inquiry committee said in its report (Appendix 6 ) our unavoidable conclusion is that allegations of scientific misconduct have sufficient substance for further investigation. They said " . . . the multiple nature of the abuses committed (improper copying and use of confidential materials, plagiarism, attempting to hide the plagiarism through suggestions of reviewers, misrepresentation of knowledge and use of citations, inconsistency regarding awareness of impropriety and actions afterward) suggest a possible pattern of misconduct and self-deception." ,* The University convened an investigation committee that provided its written report (Appendix 7) to the administration in December 1995. To come to its conclusions the committee reviewed the documents provided as attachments to the deferral. letters (see Appendix 7 for a list of these materials). It also interviewed the subject, the Researcher, the subject's department chairman, and one of the subject's colleagues. The committee made 13 findings upon which they based their conclusion. Following each finding we have included, in italics, information we found in the committee's interview summaries that we believe further explains the committee's findings.12 They were: 1) The five sections of text in the subject's 1994 NSF proposal were identical or substantially similar to those in the PI'S proposal. 2) The subject copied the text into his proposal without permission from the PI. 3) The subject failed to offset the copied text or to cite it to the source document.. 4) The committee agreed with the subject's statements to us, when responding to our inquiry letter, that the references, the copied statistical analysis text (section 5), and the copied methodology text (section 4) contained "routine" text copied "to clarify the h he University's policy states, "Conduct, inconsistent with the ethicalconduct of research and considered to be misconduct shall include: (1) serious deviation, such as fabrication, falsification, misrepresentation, or arbitrary or biased selection of data, from commonly accepted practices in proposing or conducting research or in reporting the results of I research; and (2) plagiarism or appropriating the data of another Individual and presenting it as if it were one's own. he committee i n t e r v i e w e d (A),-@), (C), and m @). In the appropriate places we have inserted the interview page number and letter. For example, A-1 represents the first page of the interview with the subject. Page 9 *) i'u- ..:.I il, .. .. .: '. " ,l 1 ..*" p'y - ..\: 'I, rationale and significance of the proposed research." However, it also concluded that section 5 contained a description'of a statistical package that "is not generally available." - The subject did not check into its availability before copying the text. The subject admitted that he did not know ifthis statistical package was even available (A-12). The committee did not comment ontthe subject's admission that he had copied sections 1-2 (in the project summary) and section 3 (the proposed significance section). I 5-6) The Researcher provided the subject with the PI's proposal with a request that he review it. .The subject understood that it was a confidential document and he did not have permission to copy it. Both the subject and the Researcher said that the subject had been given the proposal so that the subject could review selected methodolo~es(A-8, C-I). According to the Researcher and the subject, the subject had the proposal for a few hours (C-1, A-2,9) and during that time the subject claims to have photocopied 5-6 - pages @om it (A-3,9). He also said he hand writes the first versions of his proposals. Several months afler copying the pages from the PI'S proposal, he hand copied the selected passages bf text from these pages into his own proposal drafl (A-15). He told the committee that he had never reviewed a proposal but he had reviewed manuscripts and he understood such documents were conj?dential(A-3). 7-8) Three of the subject's collaborators, including the Researcher, edited the subject's proposal. Three other collaborators only read the final version. The collaborators gave their permission to be named as collaborators and they were aware of their status as such. The subject obtained the details of their research support from them. Only the three collaborators that edited his proposal were interviewed. Although all of them , testrfied to the subject's limited writing abilities (B-3,C-3, 0-2,5,6) nope of them noticed the copied passages. One of these collaborators stated that the subject had serious writing problems and his work required extensive editing and revision by others (D- ' 2,5,6). From the testimony it appears that much of the editing was devoted to the NIH submission and that one of his collaborators was not aware that the NSF proposal had been submitted fB-3). I I 9) The subject's "actions were improper, knowing and willful. This was a single incident I in his career but the time line indicates a knowing and willful:act. The material was I I copied down by [the subject] into his'grant in May 1994, at least a month before the submission of the NIH and NSF grants. He knew it was a mistake, had considered and had the opportunity to withdraw the application, but submitted the grant in any case." , The subject said thar he drd not think what he did was wrong; however he would object ifhe learned that someone had, in a similar manner, copied his text (2-21). "The 'request for exclusion of [the PI's] name as a reviewer of his NSF grant is consistent with his actions. None of his colleagues advised [the subject] that [the PI] be excluded as a reviewer." The subject testrjed that it was policy in the department to request .that members of the ~e~artment- at the major east coast university be excluded as reviewers of proposals or manuscripts (A-3,4). Other Page 10 , witnesses said that by mutual agreement, the two departments did not review each other's materials (B-5, D-I). However excluding the PI was not part of the subject's department'spolicy (B-I, C-I, D- 1). The subject said that two of his collaborators had excluded the PI because the PI had been a postdoctoral researcher in the east coast university department 0 - 4 ) . The committee was told by these people and another collaborator that they had not told the subject to exclude the PI @-I, C-1, D-I). "Lastly, [the PI's] name had not previously been on the list of reviewers he had requested not to review his [earlier NSF] grant." Although he requested that the PI be excluded as a reviewer of the NSF proposal and attributed that request to the competition between his department and the one at the east coast universify, he admitted that in prior manuscript and grant submissions he had not included this exclusion request with his submission 0 - 5 ) . 10) The subject committed a serious deviation from accepted practices and misconduct when he plagiarized the text in the PI's NSF proposal. 11) The misconduct was an isolated event. 12) The subject committed a deviation from accepted practice and misconduct in science by violating the confidentiality of the peer review process. 13) The information in the sub;ject1s proposal about his collaborators was not misrepresented. The committee concluded I that the allegations of scientific misconduct and plagiarism are well founded. There was intent to plagiarize from the grant but the misconduct did not add anything substantial to [the subject's] grant. No data w[ere] plagiarized but rather sections that appeared to improve the clarity of the text. -The committee felt that the scientific misconduct was intentional by a preponderance of evidence standard, that the misconduct was not due to honest error or honest diffei-ences in interpretation or judgments of data, and the respondent's conduct was a serious deviation from currently accepted scientific standards or from institutional policies or practices or was a violation of Federal regulations. The committee recommend that the subject be prevented from being the Principal Investigator of a Federal grant or contract for a period of three years. Furthermore, the Committee recommends he not be allowed to be a reviewer of grants and contracts for a period of three yeas. Lastly, [the subject] should inform all of his co-investigators of his misconduct in science in this three year period. Page 11 BY THE UNIVERSITY ACTIONS The Dean # received the report and fonvarded his recommendations (Appendix 8) to the Vice Pres~dentfor Research and Advanced Studies. The Dean said he ayepted the committee's conclusions. He noted that it "is not within my power to implement the recommendation that "the subject be prohibited both from being a principal investigator of federal grants or contracts" (other than those at the University) and from being a reviewer of federal grants or contracts. He endorsed the recommendation that the subject should inform his collaborators of his misconduct. He further recommended that [the subject] resign voluntarily from his position as Research Assistant Professor in the Department . . . His term of appointment is scheduled to end on August 3 1, 1996. [The subject] was given the required notice that he would not be reappointed due to the budget restraint status of the Department . . . . Because the investigative committee concluded that [the subject] was guilty of plagiarism and scientific misconduct, I will not support a request for appointment or reappointment in any department of the College . . . . beyond August 3 1, 1996. The Vice President told us that the Dean's recommendations were the final resolution to the University's case. He told us that it was likely that the subject's position would have been renewed because he received funding for the NIH proposal but, as a consequence of the investigation, the subject's position would not be renewed.13 He said the subject is required to inform each of his collaborators of this administrative sanction. Several facts we learned from the University's investigation, in combination with information we obtained during our inquiry, caused us to recompare the subject's 1993 and 1994 submissions to NIH and NSF with the PI's 1993 and 1994 submissions to NSF. Specifically, during our inquiry we had learned that the subject's 1994 NIH and NSF submissions were revisions of his 1993 submissions to NIH and IVSF. All of these proposals were remarkably similar. The copied text we identified in the subject's 1994 submissions had been used, during his revision of his 1993 proposals, to respond to reviewers' criticisms of those earlier submissions. The PI'S 1994 NSF proposal was a revision of his 1993 submission. .Both of the PI's proposals had been sent to the Researcher for peer review. The subject told the investigating comm~tteeon four separate occasions, when it asked if he had ever copied from source documents without attribution in the past, that he had not.14 l3 In a April 1'996 letter he told us that the University had co~npletedits investigation and has "determined appropriate sanctions." The subject's position had been terminated. "As a hnded investigator with a research- track appointment, he woul'd nor~nallyhave been reappointed." See Appendix 9.. '4~pecifically,he was asked (A-10) Page 12 We found these statements troubling because we saw no reason why, after writing an estimated "30" papers and submitting grants the previous year, he would have suddenly decided to rely on another's text instead of his own work. Our comparison uncovered additional sections of copied text. We found one five-line section of text (identified as section 9" in Appendix 4) that appeared to have been copied from the PI'S 1994 proposal into the subject's 1994 NIH submission. We also found three sections of text (identified as sections 6, 7, and 8, that were 6, 1, and 3 lines respectively) that 1II) So, if you've written 30 papers, what possessed you to think that at this particular time you were not up to the task of expressing yourself.......why did you take the text from somebody else's work and copy it? Have you ever done that before? 1): No On A-16 he was asked: - I'm still puzzled because you've written 30 paper and you had people helping ...... you must be ........... I don't understand. With all your other papers you've written, you've never taken any text fiom anybody else's.. .. f No Not even your own? References sometimes. But you've never copied text from a previous paper for example when you're writing an introduction? : No not copy, I just changed the words ..... you know what I wanted to express. : Why did you change it? Why didn't you just copy it down the way it was? I can't understand why you would copy this stuff down exactly the way it is and not change it at all. You wouldn't even copy ur own writing exactly the way it was. (inaudible) You would swear that if someone took all your previous work and scanned all the text against a you would never find anythlng copied down from anyone else. : Maybe, or did you not realize it was plagiarism to copy down somebody else's text? : I did not realize it. : And yet, you've never done that before? : No . . . Then you were really morally unaware of what you were doing when you did it. t : It was an honest mistake. : You still say it's an honest mistake. You had no intentions of doing anything that you knew was w ong. It was wrong. 1 knew that. You realized it at the time? Uh-huh, but I didn't think I had done nothing bad because they were my ideas. 1 4.' . . . I believe you when you say that, I don't think there's a significant difference between your ant because you added this stuff. ' I#: I have never done anything like that before. There was too much pressure, but that's no excuse (paraphrase). "sections 1-5 were found in the subject's 1994 proposals. For consistency we continued our numbering scheme for the new sections of copied text we found in the subject's 1993 (sections 6-8) an3 1994 (section 9) proposals. See Appendix 14 for a table that identifies the section number, the source document and the recipient document(s). I t I Page 13 appeared to have been copied from the PI's 1993 proposal into the subject's 1993 NSF and NIH submissions. The subject also appeared to have ,copied the 10 citations that were referenced in this copied text into his own proposals. We also found these sections 6, 7, and 8 and'the citations in the subject's 1994 submiss~onsto NIH and NSF. None of the four sections (6, 7, 8, or 9) were offset from the other proposal text nor were they cited or acknowledged to the PI. In May 1996 we wrote to the subject asking him if the text and citations in sections 6, 7, and 8 had been copied from the PI's 1993 proposal, and if so, how, since he had not been used by NSF as a reviewer, he had obtained a copy of that confidential document. We also ask if section 9 were copied from the PI's 1994 proposal. He called us and said that he had copied the text from an article by the PI; he had not seen the PI's 1993 proposal. We asked him to send us" letter stating where he had gotten the material and whether there were any other instances of unattributed copying of another's work in his proposals. In his June 1996 letter (see Appendix lo), the subject rold us that he had not seen the PI's 1993 proposal; he had copied the three sections and the citations from a review paper16published by the PI. He identified four other passages (sections 10-13) from this paper that he had copied into his proposal. He said he had copied sectior~9 from the PI's 1994 proposal. We then compared the review paper with the subject's 1993 and 1994 submissions to NSF and NIH. >*Wefound four more sections of text and citations (sections 14-17) that had been copied into these applications. We also found isolated passages and phrases that bore a I striking similarity to those in the PI's paper but did not include them in our list because we felt I they might be a reflection of the community's language for expressing commonly held concepts. Only,,$ne of the subject's four proposals (the 1994 NIH submission) contained a citation to thetPIhs,paper. In this instance, the citation accompanied a sentence in the subject's f ; explanation o hi$ responses to the reviewer's comments. It was not part of the body of the proposal. - W? We have included copies of the PI's 1993 article and the subject's 1993 NSF and NIH submissions a s ' ~ ~ ~ e n d i c e11, s 12 and 13. We have highlighted the text in sections 6-8, 10- 17 (the text copled from the PI'S-paper) and the copied citations found in the subject's 1994 and 1993 proposals consistently with the pattern established for sections 1-5. That is, the I I yellow highlighted material is common to the article (the source document) and to the 1 subject's 1993 IVSF and NIH proposals. The pink highlighted material was copied into just the 1993 NIH proposal. We have bracketed and numbered the sections in the article (6-8, 10- 17) and numbered the sections in the 1993 NIH and NSF proposals accordingly. Since much of this text was carried over into the 1994 submissions, we have highlighted, bracketed and numbered the material in these proposals, accord~ngly. However, some of these brackets are asterisked because the copied text in the 1994 proposals that was carried forward from the 1 I 1 6 ~ hpaper e was entitled, ' Page' 14 1993 proposals has been edited and is shorter than that found in the 1993 proposals. Section 9, containing text copied from the PI's 1994 proposal, has been highlighted, numbered, and bracketed similar to sections 1-5. The table found in Appendix 14 lists the 17 sections we have identified as copied from either the PI's 1993 article or his 1994 proposal. It shows the source document for each section, which of the subject's proposals these sections were copied into, and, for each section, provides the number of citations that were also copied into the subject's proposals. Of the 11 sections copied from the PI's 1993 article, 10 of them appeared in the subject's 1993 NSF proposal and 11 of them appeared in his NIH proposal. When we examined the placement of the text copied from the PI's 1993 article into his 1993 NSF proposal we found that many of the paragraphs in the "Background and Significance" section of the proposal were introduced by copied material. Similarly we found that many of the paragraphs in the "Background and Significance" section of the subject's 1993 NIH proposal, which was submitted approximately three weeks AFTER the NSF proposal are introduced by copied material. When the subject revised his declined 1993 proposals he retained all 11 sections in his 1994 NIH proposal and 6 in his NSF proposal (NSF's page limitation evidently caused the subject to delete some of the text retained in his NIH submission). The PI's 1993 paper and i 1994 proposal were the sources for a total of 17 copied sections of text, 14 of which appeared in the subject's 1994 NIH proposal and 11 in the NSF proposal. 1 Despite the subject's statement four times during the university investigation that he had not ever copied without attribution from another's work prior to the 1994 proposals, he admitted, when we confronted him with evidence, that he had copied text without attribution into his 1993 proposals. In his June 1996 response to our questions he said "Besides the above-ment~onedpassages, I have correctly cited all the other source materials used in my 1993 and 1994 proposals submitted to NSF and NIH. " However, our subsequent comparison uncovered four additional sections of copied material. Given that we disproved the subject's statements to the investigating committee, and subsequently disproved the subject's statement in his June 1996 letter to us, we do not find him a credible witness. I We deferred three allegations for investigation to the University: plagiarism, violation of the integrity of confidential peer review, and misrepresentation. The University's committee concluded that the subject had not misrepresented his colleagues' participation in his proposed research effort and we accepted this conclusion. The University's committee also concluded that, by a preponderance of the evidence, the subject committed misconduct when he plagiarized text into his 1994 NSF proposal and when he violated the integrity of the peer review process. The committee determined that the subject's actions were knowing and willful. It determined that the subject's misconduct was an isolated event. We used the investigating committee's report and conclusions as the basis for our separate conclusions about misconduct in science. Page 15 i I , . NSF defines misconduct in relevant part as "(1) Fabrication, falsification, plagiarism, or other serious deviation from accepted practices In proposing, carrying out, or reporting results from activities funded by NSF . . . " (45 C.F R 5 689.1(a)(l)). For NSF to make a finding of misconduct in science, a preponderance of the evidence must show 1) that the subject committed a bad act assoc~atedwith NSF act~vit~es and 2) that the bad act was committed with a culpable state of mind (such as willful, knowing, or gross negligence) (45 C.F.R. 5 689.2(d)). The Act - Plagiarism . . '\ . I The subject admitted to both the investigating committee and us that he had copied the identified sections of text (1-5) from the PI's proposal into his 1994 NSF proposal. He did not distinguish the copied text from original text by offset or citation to the source document. He admitted that he did not have the PI's permission to reproduce the text. We agree with the committee's conclusion that a preponderance of the evidence supports the conclusion that the subject represented the copied text in his NSF proposal as his own and that he copied the text from the. PI's NSF proposal without permission. .Additionally, we concluded that a preponderance of the evidence supports the conclusion that the subject copied, without offset or attribution, 10 sections of text from the PI's 1993 article into his 1993 NSF proposal. Six of these sections appeared in his 1994 NSF proposal. - State of Mind Plagiarism I 1 - Copying material from a source document is inherently a knowing activity. 'Our i analysis of the placement of the text copied from the PI'S published article into the subject's 1993 NSF proposal shows that the subject used text from the PI's article to guide his I composition and the structure of his background discussion. The subject selectively copied text from various parts of the paper in constructing his 1993 proposals. Despite the effort involved in carefully reading and selecting 10 different passages out of the PI's 6-page article, the subject failed to offset any of the copied text or to provide a citation to the PI's article. We believe a preponderance of the evidence supports' the conclusion that the subject's actions when he copied-text and citations from the PI's 1993 article to improve the organization and discussion in his 1993 proposals were at least knowing. 1 t The subject rece~vedthe reviews on his unfunded 1993 NIH proposal in May 1994'and his declined NSF proposal in June 1994. Although both proposals had been declined, he was encouraged by the NSF panel and by the priority score on his NIH proposal to revise and resubmit them. The reviewers of the 1993 proposals had criticized him for the organization of his methodology descriptions and the absence of a discussion on the significance of the proposed research. The subject told the investigating committee that he had the PI's 1994 proposal (the source document) for only 2 hours in early 1994 (NSF sent the proposal to the Researcher in. i Page 16 January and his review was received In March 1994). During that time, by his own admission, he selectively copied 5-6 pages (the areas of significance, rationale, methods and references) from it. His NIH proposal was submitted in June 1994 and his NSF proposal in July 1994 (approximatelj. one month, respectively, after each 1993 proposal had been officially declined). He test~fiedthat he wrote the NIH proposal in May. Therefore, approximately 3 months after he had copied the PI's proposal, while in the process of writing his proposals, the subject sought out his copy of the PI's proposal and selectively extracted the text and references he needed. For the NIH proposal he copied two blocks of text (sections 3 and 4); for the NSF proposal he copied those sections and three additional blocks of text (sections 1, 2, and 5). Given that the sections he copied from the PI's proposal are scattered throughout it (specifically pages 4, 7, 8, 16, 17, 18 and 19-20) and that the subject did not know what text or references he was going to copy until several months after he copied the PI's proposal and when he had received the reviews of his 1993 proposals and was in the process of writing his own proposals, we believe it is likely that the subject copied the PI's entire proposal and subsequently extracted text from it. The subject acknowledged to the university investigating committee that he knew he had used another's material inappropriately before he submitted the 1994 proposals, that he had the opportunity to remove the material and that he chose to submit the proposals-that contained the copied text anyway. When i*terviewing the subject, the committee said, "you had two chances, one the NIH grant and the NSF grant that you could have, and chose not to [remove the plagiarized text] in both instances." To this, the subject responded, "uh-huh . . . ." He said that before he submitted the grants he "realized that [it was] a mistake" to submit them with the copied text (A-18). Despite this, in 1994 he submitted two grants to federal agencies that he knew contained large blocks of text that had been copied without attribution to the original author. During the course of our investigation we. learned that the subject's research appointment at the University was dependent on his ability to attract research funding and that he had received notice that his position would be terminated if he did not secure outside funding1'. We learned that the subject knew that he had severe writing problems and that he relied on others to edit his material. To expedite rewriting his declined 1993 NIH and NSF proposals he referred to his copy of the 1994 PI's proposal and took the relevant text that he believed was clearly written. We conclude that the copied text improved his revised proposals because it directly responded to the criticisms the reviewers of his 1993 NSF and NIH proposals. We conclude that, in both 1993 and 1994, the subject inappropriately availed himself of the PI's written material to improve the quality of his written product. While the subject submitted cover letters with each of his NSF proposals that specifically requested that members of the Department t a major east coast University not be permitted to review the proposal, only the letter accompanying the revision 17 The inquiry committee described the subject as "career-stressed." Page 17 also requested that the PI not be a reviewer. He told the committee that his requests were consistent with department policy and were rooted in the intense competition between the two departments. His collaborators testified that it was policy to request that members of the department at the other university be excluded, but the policy did not extend to the PI, who was not a member of that department or university. We conclude that the subject included the request about the PI to avoid detection of his copied text. Similarly the subject's failure to cite the PI's publications in his discussion caused three of the six mail reviewers to explicitly draw attention to his failure and to provide citations to the PI's papers in their reviews. we' believe a preponderance of the evidence supports the conclusion that the subject willfully copied text from the PI's 1994 proposal into his NSF submission in order to increase the chances that his revised proposal would receive funding.. He copied text into his proposal that was clearly responsive to the reviewers' comments. He knew what he had done was improper and had the opportunity to remove the text but chose not to. He attempted to prevent the original author of the text from detecting the plagiarism by.requesting that he not serve as a reviewer of the proposal. J ' The Act - Use of an NSF Proposal Received in Confidence t The subject told the investigating committee and us that he knew that the PI's 1994 proposal had been sent by NSF to a member of his department's faculty with a request for review. The Researcher erred when he asked the subject's opinion of selected methodologies in the proposal without first consulting the NSF program manager: The subject,admitted to copying pages from the PI's proposal and subsequently using the text in his submission to NSF. Therefore, we conclude that a preponderance of the evidence supports the finding that the subject violated the confidentiality associated with NSF's peer review process when he copied from an NSF proposal provided to him by the Researcher who "stressed [its] confidential nature. . . .1118 State of Mind - Use of an NSP Proposal Received in Confidence The subject and the Researcher told both the committee and us that the subject had been told, when he received the PI's proposal from the Researcher, that it was a confidential document. Although the Researcher failed to obtain NSF's prior permission, in accordance with NSF's explicit instructions about confidentiality, he orally told the subject the proposal was confidential. The subject said he would object if someone had copied his material without permission. Despite the Researcher's explicit instruction and his own understanding of confidentiality, the subject made a copy of the PI's proposal and, several months later, used text from it in his own submissions. We conclude that the subject was aware of his duty to maintain the proposal in confidence and ignored that duty when he plagiarized from it. A preponderance of the evidence supports the finding that the subject acted, minimally, ' * ~ ~ ~ e n2,d i x Page 18 knowingly when he used text from a document he received in confidence, and knew was confidential, in his own proposal. Our Conclusion about the Plagiarism in the NSF Proposals We conclude that the subject committed plagiarism when he knowingly copied text 10 passages of text from an article published by the PI into his ,1993 NSF submission. We further conclude that the subject committed plagiarism when he willfully copied 5 passages of text from the PI's confidential NSF proposal into his own NSF submission. He violated the . confidentiality of peer review by knowingly photocopying pages from a proposal he had been instructed was confidential and later copying selected passages into his own NSF submission. His actions were serious deviations from accepted practices in the scientific community, hence misconduct in science. j \ Under 8 689.2(b) of NSF's misconduct in science and engineering regulation, when deciding what actions are appropriate when misconduct is found, NSF officials should consider the seriousness of the misconduct, the intent with which the subject acted, any evidence of a pattern, and finally its relevance to other funding requests or awards involving the University or individual. Evidence of a Pattern of Plagiarism The University's inquiry committee thought that the subject's actions might be "suggest[ive of] a possible pattern of misconduct and self-deception." However, the investigating committee concluded that the subject's plagiarism "was a single incident in his career" on the basis of their finding that the "misconduct appears to be an isolated event." From important evidence that became available to us after the University submitted its investigation report, we concluded that the subject exhibited a pattern of plagiarism. We found that the 1994 NSF proposal contained more sections of plagiarized material than the NIH proposal and we found, on further investigation, that the subject had lied to the committee when he said, four separate times, that he had never plagiarized in the past. We found 10 sections of text that had been plagiarized from a 1993 article by the PI into the subject's 1993 NSF and NIH proposals (the NIH proposal contained one additional section). He was not truthful with us in his June 1996 letter when he stated that, except for the passages he had identified, all other source materials were correctly cited. Our subsequent review found four additional passages of text plagiarized from the PI's article. We conclude that the subject exhibited a pattern of plagiarism. The first instance of plagiarism that we are aware of occurred in September 1993; when he submitted his NSF proposal, the second instance occurred when he submitted his NIH proposal, almost 1 month Page 19 / later. The third instance occurred in June 1994 when he submitted his 1994 NIH proposal. The fourth instance occurred about 2 weeks later when he submitted his 1994 NSF proposal. Seriousness w e Jbelieve the following facts support the conclusion that the subject's actions are far more serious than a single instance of plagiarism caused by frustration or time pressures. In this case, the subject did not simply plagiarize text and citations into one proposal. Over the period of 11 months and on four separate occasions he plagiarized multiple passages of text into proposals he submitted to two federal agencies. Since each proposal iterated on the last, he edited or retained the plagiarized material from his most recently submitted version into his current version and added additional plagiarized material. The size of the plagiarized passages ranged from 1 or 2 lines of the proposal to 22. Many of the shorter passages were found as the introductory sentence to paragraphs while the longer .passages represented the entire content or substance of a proposal section. The plagiarized passages were variously found in the NSF Project Summary, the Background, the methodology in one Specific Aim, and Statistical Analysis of Quantitative Data While we could not identify a specific impetus for the subject's plagiarism in the 1993 proposals, the material plagiarized into the two 1994 proposals waS clearly a response to the 1993 proposals' reviewers' comments. In response to our inquiry letters the subject said "it was not my intention to steal his ideas, methodologies, or innovative concepts . . . [algain the sole purpose was to express in a clear, concise manner ideas we held in common." However, as the investigating committee found, his statement is made false by his admission that he plagiarized text (section 5) describing statistical packages that he did not have and that were not available to him. His plagiarism is made more serious because it was not simply copying well-expressed but commonly held ideas, but copying this novel information and idea from the PI'S proposal. The subject's misconduct is made more serious because he lied to the investigating committee on four separate occasions when, in response to questions, he falsely claimed that he had never plagiarized in the past. In addition, in'1994, the subject requested, in a cover letter to his proposal, that the PI not be used as a teviewer. The evidence supports the conclusion that this request was not, as the subject .claimed, in conformance with department policy. Instead, we agree with the University inquiry committee's conclusion that this request was'an attempt to prevent the original author from detecting the plagiarism. We believe the scientific community has provided a separate statement about the seriousness of the subject's actlons. On the basis of the case we deferred, the investigating committee recommended that the subject be barred from serving as a principal investigator on federal and contracts for 3 years and that he be prohibited from serving as a reviewer of these documents for the same 3 year period. Page 2 0 Recommended NSF Action In settling its portion of this case the Office of Research 1ntegrity19 confined its considerations strictly to the plagiarism of two sections of text (sections 3 and 4) into the 1994 NIH proposal. ORI did not consider the evidence of a pattern presented by the plagiarism in the 1994 NSF proposal and did not discover the plagiarism in the 1993 proposals or the additional section (section 9) in the 1994 NIH proposal. Further, sections 1, 2 and section 5 (which consisted of plagiarized text that contained ideas and concepts unknown to the subject) are found only in the 1994 NSF proposal. ORI did not evaluate the violation of NSF's confidential peer review process or the subject's attempt to prevent the original author from finding the plagiarism by requesting that he not be used as a reviewer. We believe the circumstances of the NSF misconduct case are more serious than those considered by the O N case. We believe that NSF should take action to protect the Government's interest in funding research. We recommend that NSF's Deputy Director take the following actions to protect the government's interests: (1) NSF should send a letter of reprimand to the subject stating that it has concluded that he committed serious deviations from accepted practices and thus misconduct in science when he violated the confidentiality associated with NSF's peer review process and plagiarized text from a confidential NSF proposal and a publication into his own NSF proposals.20 (2) NSF should debar the subject from receiving government grants for a period of 2 years from the date of the final disposition of this case.21 We consider this recommendation particularly important in protecting the Federal government's interests because on two separate occasions the subject submitted proposals containing plagiarized material to two separate federal government agencies. Hence, he shows a pattern of disregarding ethical practices and a lack of present responsibility when applying for Federal funds. (3) IVSF should require that for two years following the end of the 2-year debarment period, when the subject is a principal investigator or co-principal investigator on a proposal submitted to NSF, the subject will certify in writing that he has recently reviewed NSF's Misconduct in Science and Engineering regulation (45 C.F.R. 5 689), "In ON'S Voluntary Settlement Agreement with the subject, the subject agreed that he coinmitted misconduct in science by plagiarizing material into his NIH proposal. For 3 years'he will certlfy in every PHs application or report that he has cited and acknowledged his contributors. His certification will be endoised by an institution official. He is excluded from serving in any advisory capacity to PHs. After executing the agreement OR1 informed the subject that his name would be maintained on the ALERT system for a period of 3 years. 2 0 ~ h iiss a Group I action (see 8 689.2(a)(l)(i)). 2 1 ~ h iiss a Group I11 action (see $689.2(a)(3)(ii)). Page 21 .. that the proposal is free of any misconduct, and that.the proposal has been reviewed as described below.22 i (4) NSF should require that for two years following the end of the 2-year debarment ' ' period, when the subject is a principal investigator or co-principal investigator on a proposal submitted to NSF, the subject will ensure that his department chairperson, or equivalent, has signed an assurance stating that, on the basis of that person's reading of the proposal and to the best of that person's knowledge, the proposal does not contain any plagiarized material and all the source documents have been appropriately cited. I , . . 7 (5) T h e subject should be requested to send his certification, as required in (3), and his department chairperson's assurance, as required ,in (4), to the Assistant Inspector General for Oversight in NSF's Office of Inspector .General, for retention in that . Office's confidential file on this matter. r (6) NSF should prohibit the subject from participating as an NSF reviewer, advisor or consultant for a period of 3 years from the date of the final disposition of this case.23 rr We fonvarded the draft investigation,report to the subject for comment and received a , response from his attorney on September 26, 1996 (see Appendix 15). We reviewed the response ahd concluded that it did not contain any additional information that altered our conclusioris about the subject's actions. Z . I The attorney stated24 that the subject had fully cooperated with "all aspects of the investigation process." This statement is, however, refuted by our finding that the subject lied on four separate occasions when the university committee asked him if he h,ad plagiarized in the past: Our subsequent investigation discovered additional substantial plagiarism in both \ 1993 proposals and a new passage in one of the 1994 proposals. The subject also told the investigating committee that his request that the PI not be used as a reviewer was consistent with department policy. None of the department members interviewed, including the department chairperson agreed that this was a departmental policy. We concluded his request was simply an attempt to prevent the original author from reviewing the subject's proposal that contained text plagiarized from the author's review article and proposal. The attorney also stated2' that the subject's inisconduct is "a single episode repeated throughout'his several efforts" and that the subject's actions did not represent a pattern of plagiatism. The facts of the case, which are not in dispute, show that the subject sequentially . i 22 This is a rob^ I1 action (see $689.2(a)(2)(ii)). 2 3 ~ h iiss a Group I11 action (see $689.2(a)(3)(iii)). 24 Paragraph 2 page 1 of the response. - , "paragraph 4 page 1, paragraph 2 page 2, and paragraph 4 page 3 of the response. Page 22 submitted four proposals and .that each proposal contained new sections of plagiarized material. The attorney claimed26 that the investigations into the subject's misconduct had negative effects on the subject's career. We subsequently inquired about the subject's current - - em~lovment status and learned that his position at the university had been terminated on August 3 1, 1996. However, effective September 1, 1996, University -(the consultant group) became the subject's employer. By an annually - - renewable cdntract with the university, the subject serves &ithe principal investigator on the NIH grant and the university reimburses the consultant group 100% of the subject's salary support from that grant. The subject's current salary negotiated in the contract includes a 4.2% salary increase over the previous year. The university remains the grantee for the award and retains all the indirect cost recoveries. Further, the subject has retained his on-campus research laboratory space. We were told that the subject's pending application to be a volunteer unpaid faculty member has the full support of the subject's former department chairman. We do not find that the evidence supports the attorney's claim. The attorney's response requested that we modifjr our conclusions to be consistent with those of the more limited case considered by ORI.~' However, the case considered by NSF is more serious than that considered by ORI, although the cases do have some evidential overlap. ORI's case was limited to the plagiarism in the subject's 1994 NIH proposal. ORI's settlement required certifications and assurances from the subject on the basis of the plagiarism in it. The case resolved by the university was broader than the case before OR1 because the university considered the plagiarism in the subject's 1994 NIH and NSF proposals as well as the violation of confidential peer review. Based on that evidence the university investigating committee recommended that the subject be debarred from receiving government funds and participating in peer review for 3 years. The Dean accepted the committee's coriclusions but noted that he was powerless to implement the debarment and the peer review recommendations. We agree with the university investigating committee's evaluation. Our case is distinct from and more extensive than the university's case and far more extensive than the case before ORI. We considered the plagiarism in the subject's 1993 and 1994 NSF proposals, the breach of confidentiality in NSF's peer review process, and the evidence of a pattern of plagiarism, exhibited by the evidence of new acts of plagiarism found in each of four sequentially submitted proposals, the 1993 and 1994 NSF and NIH proposals. NSF's misconduct regulation specifically contemplates that the sanction imposed be based, in part, on the evidence of a pattern of behavior. Accordingly, we believe the actions we have recommended are reasonable given the subject's misconduct. Because the subject submitted proposals containing plagiarized material to two separate federal agencies on two separate 26~aragraph 3 page 3 of the response. 27 Paragraph 4 page 3 of the response. . Page 23 , occasions and breached confidential peer review, our conclusion is that the subject's conduct warrants a severe action, debarment. Our debarment recommendation is consistent with previous decisions by NSF in serious plagiarism cases.28 The attorney requested that the investigation report be restricted to only NSF issues and therefore the government-wide debarment action be deleted.29 We have explained in our report and outlined above why we believe the subject lacks present responsibility and debarment is warranted We noted that, in its more limited case, the university investigating committee's recommended debarment period is longer than ours. The federal debarment system gives each agency's debarring official the authority and responsibility to enforce the policy of the federal government to conduct business only with responsible persons.30 When presented with evidence of a person's lack of present responsibility, it is the debarring official's duty-after considering the seriousness of the person's acts and any mitigating factors-to take action to protect the interests of the federal government as a whole.31 To deal with cases in which more than one agency has an interest in a proposed debarment-as in the instant case-agencies are encouraged to coordinate their debarment actions.32 Currently, the subject receives no NSF funds but the subject does have an active NIH award (see footnote 3 for the specifics on this award). Consequently, the Deputy Director may want to coordinate ! ! - m ith the h Director of t the Institute that manages the subject's award. The Deputy Director may also want to provide a confidential copy of our investigation report and its attachments to the Director of the Institute. 2 8 ~ e ee.g. , cases M90110044, M92020007, M93020006. 2g~aragraph 4 page 3 of the response. 3045C.F.R. 5 620.115(a). "45 C.F.R. 54 620.115(a),-.300. Page 24
Peer Review violation Plagiarism (Verbatim)
Published by the National Science Foundation, Office of Inspector General on 1997-10-21.
Below is a raw (and likely hideous) rendition of the original report. (PDF)