oversight

Data Tampering / Sabotage / Fabrication

Published by the National Science Foundation, Office of Inspector General on 1999-05-05.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                           Closeout for M97020004
This case was brought to the attention of OIG) January 1997. A program manager1
provided our office with a letter she had received from a ~cientist.~The letter
described allegations of fabrication against a graduate student3and the university's
actions to resolve the matter.

OIG's investigation report and NSF's Deputy Director's 9 February 1999 letter
describing his decision constitute the closeout for this case.

Cc: Integrity, IG




                               Page 1 of 1                           M97-04
                       NATIONAL SCIENCE FOUNDATION
                           4201 WILSON BOULEVARD
                          ARLINGTON, VIRGINIA 22230


                                      February 9 , 1999


    OmCE OF M E
   DEPUTY DIRECTOR


 CERTIFIED MAIL-RETURN RECEIPT REOUESTED




 Re:   Notice of Misconduct in Science Determination
 Dear Ms.
 The National Science Foundation's Office of Inspector General
 (OIG) issued an Investigative Report on October 1, 1998 in which
 it concluded that you fabricated research data in your Ph.D.
 thesis. A copy of the OIG investigative report is enclosed.
 J'4isconduct in Science and Proposed sanctions
Under the National Science Foundation's (NSF.) regulations,
umisconductuis defined to include "fabrication, falsification,
plagiarism, or other serious deviation from accepted practices in
proposing, carrying out, or reporting results from activities
funded by N S F . " 4 5 CFR. §689.1(a).
The Foundation's administrative record indicates that YOU were
 formerly a doctoral student in the           ~e~artment at the
'University of                                      you submitted
a thesis to ob                  Your thesis research was S U D D O   ~ ~ ~ ~




 In 1996, allegations arose that the measurements in two chapters
of your thesis were fabricated. A University Committee
investigated the allegations and concluded that you fabricated
research data by cutting and pasting          to eliminate actual
data and add new           features (Ex
                                      W I I
                                          I   N    The Committee
concluded that tite heart of [the] dissertation Iwasl based on
fraudulent datai1and it found "a very clear pattern that
undermines the entire basis for the research reported in the
dissertation." (Exhibit 1A). You did not contest the majority of
the allegations and withdrew your thesis. The University
rescinded your Ph.D. degree in February of 1997. The University
also took additional measures to notify the appropriate
institutions of the research fabrication.
  NSF's Office of Inspector General conducted its own investigation
  and agrees with the Committee that you fabricated research data.
  OIG concludes that you deliberately fabricated the data and that
  you committed a serious deviation from accepted practices.
 I concur with the Committee and the OIG1s conclusion that you
 fabricated data in your Ph.D. thesis. The record demonstrates
 that you deliberately fabricated the data by cutting and pasting
 spectra. Your fabrication of significant research data in your
 Ph.D. thesis is a serious deviation from accepted practices
 within the scientific community. I conclude that you committed
 misconduct in science.
 In deciding what action is appropriate to take in response to the
 finding of misconduct in science, NSF has considered the
 seriousness of the misconduct, whether it was deliberate or
 careless; whether it 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. See 4 5 C.F.R. §689.2(b).
 I am issuing this finding of misconduct in science and letter of
 reprimand to express strong disapproval of your conduct in this
 matter. Research fabrication is a serious offense because it
 distorts the scientific record. The scientific record is the
 foundation for all scientific research.
 In determining the appropriate sanction, however, I conclude that
 further action is not necessary to protect the Government's
 interest becau6.e theuniversitv took numerous stews to address
 the fabrication and ou have azvised NSF that you-have not worked
 in the field of -since          you forfeited your degree.
 Nevertheless, if you submit any research proposals or reports to
NSF or report on the results of NSF-supported research within
 three years from the date of this letter, you must submit a
separate certification to NSFts OIG. The written certification
shall state that to the best of your knowledge, the documents
contain no fabricated or falsified data. The certification
should be sent to the Associate Inspector General for Scientific
Integrity, 4201 Wilson Boulevard, Arlington, Virginia, 22230, at
the same time that you submit the research proposal or report to
NSF or report the results of NSF-funded research. In addition,
the supervisor or principal investigator of the project must also
submit an assurance to the OIG that to the best of his or her
knowledge, your research proposal or report submitted to NSF, or
report of results from NSF-funded research does not contain any
falsified or fabricated data.
Procedures Governins Ameals
Under NSF1s regulations, you have 30 days after receipt of this
letter to submit an appeal of this decision, in writing, to the
Director of the Foundation. 4 5 CFR §689.9(a).. Any appeal should
be addressed to the Director at the National Science Foundation,
4201 Wilson~Boulevard,Arlington, Virginia 22230. For your
information we are attaching a copy of the applicable
regulations. If you have an questions about the foregoing,
please call Lawrence Rudolph, General Counsel, at (703) 306-1060.'




                                Joseph Bordogna
                            C/Acting Deputy Director

Enclosures (21
Misconduct in Science Regulations
Investigative Report
      Confidential




Office of Inspector General

   Investigation Report

  OIG Case M97020004


     1 October 1998
   REPORT OF INVESTIGATION INTO AN ALLEGATION OF MISCONDUCT IN
                     SCIENCE AND ENGINEERING


                                            SUMMARY

           The Office of Inspector General (OIG) has
  subject), formerly a graduate student at the University of
  fabricated data in her Ph.D. thesis. This cohclusion is based on investigations by the
  institution and OIG. OIG recommends that NSF make a finding of misconduct in science
  and send the subject a letter of reprimand notifying her that NSF has made that finding.

                                         BACKGROUND
            The subject received her PhD. in               from the institution on-
~               e thesis research
                            r     was supported in part by an NSF award' After she left the
 d t u t i o n , she asked another student to bring her a box that she had left behind in her
 Ph.D. advisor's laboratory. A postdoctoral student, who had been attempting without
 success to reproduce the subject's measurements, discovered evidence of fabricated
 research data in the box.2 The evidence was reviewed by the
                                                     (the Committee)
                                                      ~ e a n ) .The
                                                                  ~ subject did not contest the
                                                       her, and as a result, returned her Ph.D.
 degree and withdrew her thesis. The subject also apologized to the College of Arts and
 Sciences, the D e p a r t m e n t , and her 'Ph.D. advi~or.~        The subject's degree was
rescinded by the institution-no
       The subject's Ph.D. advisor informed ;his NSF program officer about the matter
and, at the program officer's request, summarized the incident and the institution's
response in a letter. That letter was forwardedito OIG at the request of the institutioa5




     disputed. Letter fiom the subject to Associate Dean at 1 (1 October 1996) (exhibit 1C); letter
     fiom the subject's Ph.D. advisor to Chair of the j~ornrnitteeat 1 (25 June 1996) (exhibit 1B).
3
  Letter from Committee to Associate Dean (25 June 1996) (exhibit 1A).
4
  Letter from the subject to Associate Dean at 2 (1 October 1996) (exhibit 1C).
' The letter, dated 15 November 1996, was forwarded to OIG on 24 January 1997. Pursuant to a
+                                                                                                     -
    subpoena issued 25 March 1997, OIG requested and obtained copies of documents relating to
    the institution's investigation into and actions taken in response to the allegation that the
    subject had fabricated data in the course of1 her thesis research. During the course of
    reviewing the material provided by the institution, OIG requested and received additional
    documents: on 5 June 1998, a copy of the Subject's thesis and, on 20 August 1998, the
    original documents that were attached to the 25 June 1996 letter fiom the subject's Ph.D.
    advisor to the Chair of the Committee.
                                                  I
                                                             I




                           THE INSTITUTION'S INVESTIGATION
           According to the subject's Ph.D. advisor, chapter two of the thesis reported the




    presented in ch ter two of the thesiskere kabricated.* None ofthese measurements
                    9
    were published. The subject's Ph.D. advisor also alleged that one measurement in
    chapter three of the subject's thesis was fabriqated and that this fabricated measurement
    was published.'0                               I
                                                   ,
            The Committee that reviewed the evidence of fabrication was composed of five
    faculty members f b m the Department of            The Committee recommended that
    the subject's Ph.D. degree be rescinded.    I
                                                         I



          The Committee explained that in their r&ew of the case:
                                                         !
          [The Committee] learned that the key chapter of the dissertation[, chapter
          two,] involved essentially three kinds of measurements . . A thorough..
          review of the results of this chapter revealed t h a t             and .)
         r e p o r t e d in the thesis had been altered through cutting and pasting
          to eliminate actual data and add new spectral features where none
          appeared in authentic spectra. Fabri4tion of-spectral         results was
          also found in a second chaptea, chaptkr three,] that has been published.
                                                     I
                                                     I
6The catalyst is

'~etterfrom the subject's m o - ~ & e                                  at 1    &
                                                                               June 1996) @,bit
    .                 two of the thesis also included a study of the kinetics of one of the
                        ctions. The subject's Ph.D. advisor also alleged that some s p e c t r a
     with no thesis counterparts were fkbricated and used to make presentation overheads
     illustrating how the kinetics of               ' ere followed. Id. at 1-2. Evidence of this
     fabrication fimn the box the sub=is                available for review in our administrative




     NSF support.
1




    479. The author informed OIG that this paper ~           l belcorrected.
                                                                 I

             The rest of the work in that chapter wad carried out by [a collaborator] and
             is in order. Our review leads us to the kery clear conciusion that the heart
             of [the subject's] dissertation is based dn fraudulent data These instances
             are not isolated, but constitute a very/clear pattern that undermines the
             entire basis for the research reported in h e di~sertation.~"~
                                                                 I
                                                                 I

              The Associate Dean notified the subjdt of the allegations against her, and after
     discussing the allegations with her Ph.D. advisfr,I2 the subject called the Associate Dean.
     The subject told him that she had decided to ,return her Ph.D. thesis and degree to the
     institution, and requested a copy of her Ph.Dj advisor's letter detailing the allegations.
     The Associate Dean provided the requested document and asked the subject to respond to
     each of the allegations."
                                                             I
             In her response to the Associate Deap, the subject did not contest her PhD.
     advisor's allegations that spectra for ~ e t e m + a t i o n and-analyses
                                                                  s            reported in
     chapter two of her thesis were fabricated. The subject did, however, contest the
     allegations concerning the 0measurkments reported in chapters two and three
     of her thesis. In the last paragraph of her respobe, the subject said:
                                                             I

        +
                    In conclusion, I will not contbt the majority of charges filed
            against me by [my Ph.D. advisor]   regs-g      [clhapters [t]wo and [tlhree of
            my Ph.D. [tlhesis. As a result, I am p m i n g the degree of Doctor of
            Philosophy to the [institution] and yithdrawing my thesis h m the
            College of Arts and Sciences. I loo4 forward to putting this incident
            behind [me] and wish to thank you [personally for all the help and
            guidance. you have given me during the' recent events. I apologize to the
            college of Arts and Sciences, the e p a r t m e n t , and to [my Ph.D.
            advisor] for tarnishing the reputation of the institution^.'^


                                                         i




                     -
         The institution rescinded the subject's t h . ~ degree
                                                             .  -no                      The
institution's registrar also took steps to ensurelthat all transcripts issued for the subject
would include a letter stating: "To Whom It h a y Concern: This student's degree was
revoked for academic disciplinary reasons. duestions regarding this student may be
directed to the office of the Dean of The ~ o l l e ~ e . " ' ~

       -no               the Vice Provost fAr Research and Graduate Affairs (Vice
Provost) sent a memorandum to the institution's administrators, the Chair of the
Department of              and members ok the subject's dissertation committee
informing them of the rescission of the subj'ect's Ph.D. degree. The memorandum                    -
                                                     I
II
   Letter from Committee to Associate Dean at 1 (25 Itme 1996) (exhibit 1A)
12 The subject discussed the allegations with her Ph.?. advisor a                            L
13
14
    which was held at the University of
                                        7      during )he week o
   Letter from Associate Dean to the subject I6 A u y t 1996).
  Letter from the subject to Associate Dean at 2 (1 Optober 1996) (exhibit 1C).
IS
  Memorandum from University Registrar to Vice Provost (12 March 1997).
                                                    I
     directed that, if a letter of recommendation had been written on behalf of the subject, or if
     any organization had been contacted about the laward of the subject's Ph.D., each person
    and organization previously contacted should lbe notified that the subject's degree had
    been rescinded. l6 The Vice Provost also fo-ly            notified all organizations that had been
    informed of the institution's conferral of the Fbject's degree that the degree had been
    rescinded. The Provost notified the presiden) of the university where the subject had
    taught of the rescission of her degree." The ProvostLinformed the subject of the
    institution's action rescinding her degree and ofithenotifications of that action sent by the
    institution to specific individuals and ~ r ~ a n i z a t i ~ n s . ' ~



             01G obtained from the institution copies of documents relating to the subject,
    including the subject's student reconis, and copies of documents relating to the
    institution's investigation into and actions takb in response to the allegation that the
    subject had fabricated data in the course of per thesis rescaroh. OIG also obtained
    original documents from the box left behind by the subject." On 26 August 1998, OIG
    sent the subject copies of documents obtained from the institution and asked for her
    comments on the allegations and any addition4 information she wanted to provide. The
    subject said:
                                                       I
          My official response regarding the           I
                                                       specifics
                                                               of allegations of data
          fabrication in the course of my doctoral research was outlined in my letter
          to [the Associate Dean] [of 11 October 1996. Thus, I will not restate it
          here other than to say I did not contht the majority of charges when
          presented to me by [my Ph.D. advisor] y d do not now. I would, however,
          like to take this opportunity to express my deep regret for the situation, I
          alone, have created. After much reflktion, I am truly ashamed of my
          actions which have tarnished the repu4tion of a professor, a department,
          and a university along with breaching the trust of those closest to me.?'


        NSF defines "misconduct in scienck," in pertinent part as "fabrication,
falsification, plagiarism, or other serious deviation from accepted ractices in pmposing,
canying out, or reporting results fiom activitiel fimded by NSF." P The subject in this
                                                                            '
case was a graduate student, who received a stipend from her Ph.D. advisor's NSF award
during the course of her thesis research. Her research supplies and expendables were
I
  %emorandurn from Vice Provost to Acting Dean i f The College, Associate Dean,Chair of the
    Department of                  the subject's Ph.D. advisor, and members of the subject's             -
    dissertation
 17
   Letter fiom Provost to President of -university          (5 March 1997).
                                                     I
 "Letter from Provost to the subject (19 March 1 9974.
I
  9These documents were attached to the 25 June 1996 letter (exhibit 1B) from the subject's Ph.D.
    advisor.
20
   Letter from the subject to OIG (1 September 1998) (emphasis in original) (exhibit ID).
"45 C.F.R. § 689.2(a)(l).
  charged to that award. There can be no doubt that fabrications that undermine the basis
  for research reported in a graduate student's thesis and lead the student to accede to the
  rescission of a conferred Ph.D. degree are a lserious deviation from accepted practices
  and, therefore, misconduct in science under NSF's regulation.

         Below, we present the evidence thit supports our conclusion that t h e m
  determinations, s p e c t r a , and a l F e s were fabricated.

          ~abricatedm
                                                     I
                    eterminations in Thesis Cha~terTwo
                                                    I
         High pressure liquid chromatography @LC)           is used to separate multiple
  chemical compounds that are dissolved in sblution. HPLC instruments consist of a
  reservoir of solution, a pump, an injector, a lseparation column, and a detector. The
  compounds in solution are separated by injecting a sample fiom the reservoir onto the
 column. The diffkent compounds in solution $ass through the column at different rates,
 and the detector records the diffkrent compounhs as they come out of the column. HPLC
 instruments produce chromatograms, graphs? showing the compounds corning off the
 column over time; the peaks on the chromatograms correspond to the retention times, the
 time at which each different compound comes but of the column.
          Exhibits 2A, JA, and 4A are figures 1./5,2.16, and 2.17 from chapter two of the
  subject's thesis. These figures are HPLC chromatograms that purportedly demonstrate
 meterminations for certain chemical reactiobs. Exhibits 2B, 3B, and 4B are copies of
 documents h m the box left behind by the bbject that demonstrate that the subject
 fabricated these HPLC chromatograms for her fhesis." In the chromatograms of exhibits
 2B, 3B, and 4B,cutting and taping is apparent in the graphs and in the numbers below the
 graphs. The graphs in these chromatograms lare composed of several small pieces of                     I




 paper taped together. The pieces of papa haye been carefhlly matched at the edges so
 that the resulting graph appears to be a single continuous line.
                                                  I
                                                  I
         The shapes of the graphs in the chrom~togramsof thesis figures 2.15, 2.16, and
2.17 are identical to the shapes of the graphs in the cut-and-taped chromatograms,
although some of the numbers (graph labels ar/d numbers below the graphs) in the thesis
figures are not the same as in the cut-and-taped chromatograms. The identical shapes of
the graphs are significant because random feahes in chromatograms, such as noise on a
graph's baseline, will not be the same in twb different c h m m a t o ~ - - e v e nin two
chromatograms produced by repeating exactly the same analysis. If the same random
features occur in the shapes of the graphs in hJo chromatograms then the same graph was
used in both. The peaks on the graphs in the dut-and-taped chromatograms are identical

%e vertical axis on these graphs is time (inCreasidgfkom top to bottom).                            -
q w o copies of each document are included in ex$bits 2B,3B, and 4B: the first copy was made
  with enhanced contrast, and the second, with normal contrast. The enhancedcontrast copies
  of these documents allow the cutting and taping lin the original documents, which are available
  for review in our administrative record, to show clearly in the copies attached to this report.
  The handwritten notation on the document in Lxhibit 2B, "same as thesis fig p. 68 except
                                                  I
  retention time changed[,]" was added by the subject's    Ph.D. advisor after the document was
  removed from the box.
     to the peaks in the three thesis figures; modver, the very small "noise" features on the
     baselines on the graphs in the cut-and-taped dhromato-s         are identical to those in the
     three thesis figures. Even though all of the   ribbers  in the chromatograms in the thesis
     figures are not the same as those in the cut-and-taped chromatograms, OIG concludes that
     both chromatograms contain the same fabricatkd graph.
             The subject does not contest the charge that the =determinations         in chapter two
     of her thesis were fabri~ated.'~
            Fabricated        Soectra in Thesis Chapter Two
                                                      I
                                              dPectroscopy is a method used in chemistry
  to provide unique infonnation about molecules! The d i f f m t peaks in an-   spectnim
                                               ).
  comspond to resonances in the molecular stru~ture of the sample analyzed. Analyses of
                                                    9
 s p e c t r a depend on recognizing patterns the peaks of the spectra.
         Exhibit 5A is figure 2.5 from chaptd two of the subject's thesis, which is an
 e               c purportedly demonstrating that she had obtained a successfid result h m
  a I                                               experiment. Exhibit SB is an
  spectrum from the box the subject left behind 'that is the product of culling and taping.
  Spectral features in exhibit 5B were created by taping into place small pieces of paper
 with peaks and eliminated by taping small *es          of paper with noisy baselines over
 existing peaks. The spectral features have been canfilly matched at the edges of the
 small pieces of paper so that the resulting spectrum appears to be a single continuous line.
 Although exhibit 5B is not itself h m the thedis (it appears to be an overhead slide that
 would be used in an oral presentation), thy fabricated spectrum in this document,
 including peaks and noisy baselines, is identical to the spectrum in thesis figure 2.5."
        The subject's PLD. advisor alleged thai                spectra in chapter two of the
subject's thesis were the result of "doctoring"l and that -resonances        were added to
                             The subject contes!ed the charge that the r e s o n a n c e s in
                           were additions to                     spectra. She said: "[tlhe
original spectrum is that of the substrate do                     The corresponding-
spectra are that of two separate substrate pools and thus lead to the differences in
chemical shift of the                  products.i" While the subject's explanation may
accurately describe the origin of pieces of s p e c t r a that she used in creating exhibits
5A and 5B, OIG does not believe that                      explanation does or can justify
creating       spectra by cutting and taping

24
   Letter fmm the subject to OIG (1 September 1498) (exhibit ID); letter from the subject to           -
    Associate Dean at I (1 October 1996) (exhibit I?.
25
  An enhanced-contrast copy of the document and a normalcontrast copy are included in exhibit
     SB.
26
   Additional evidence of fabricated-       spectra d m the box the subject isfi behind is available
27
28
    for review in our administrative record.
  Letter fiom the subject's P ~ . Dadvisor
                                   .              o d ~
                                           to chair      committee (25 June 1996) (exhibit LB).
                                                     I 1996)
  Letter from the subject to Associate Dean (1 October        (exhibit 1 C).
             Having considered the evidence fro4 the box the subsect lefi behind and the
     subject's response to the allegation, OIG concludes that the    J  spectra in chapter
     two of the subject's thesis were fabricated.
           Fabricated s p e c t r u m in Thesis chapter Three that Was Published
           Exhibit 6A contains figure 3.6(d)      fro+        chapter three of the subject's thesis, an
   -spectrum          that pluportedly demonstrates a successful result in certain
    experiments. The s i g n i f i c a n ~ r e s o n a n c ehthis figure is the most prominent feature
    in the                 between -1 7 and -1 8; (his resonance has a peak extending below
   the baseline closely associated Mth a peak extending above the baseline. The same
   spectrum was also published in two papers, d shown in exhibits 6 8 and 6 ~ We               : will
                                                                                                   ~
                                                            I
   mfer to this spectnun as the published spectrum.
                                                            I
           Exhibit 6D is an s p e c t r u m , fbml the box left behind by the subject, that is
   the product of cutting and tapinggJOExhibit 6b was created by taping eight small pieces
   of paper containing diffkrent spectral featuresI including resonances and noisy baseline
  segments, to a blank sheet of paper. A labelq axis has been taped below the spectrum.
  The spectral features have been carefully matched at the edges of the small pieces of
  paper so that the resulting spectrum appears to1be a single continuous line. Although the
  published spectrum is not identical to the cut-and-taped spectrum of exhibit 6D (for
  example, the same prominent feature appearing near -9.8 in the published spectrum
  appears near -9.3 in the cut-and-taped spec+),                the published spec-      incorporates
  some of the cut-and-taped features. OIG believes that any spectrum that incorporates cut-
  and-taped features, without explicitly acknowltdging such manipulation, was fabricated.
 The noisy baseline between -18 and -20 in the published spectnun is identical to the cut-
 and-taped-spectnun, which in this region is composed of two smaller pieces. The
 published spectrum and the cut-and-taped spytrum are also the same between -1 1 and
 -17, a region that includes a prominent feature along with noisy baseline. The cut-and-
 taped spectrum in this region, too, is composed of several smaller pieces.

       The subject rejected her PhD. advisorstCharge that the I )resonances in these
spectra were fabricated, arguing that the y g e d l y fraudulent published data "was
observed at the reported chemical shift during one experiment. This result was never
reproduced by myself."'    OIG believes that obsewhg the desired effect in one of her
experiments does not and cannot justify the subject's misrepresentation of spectra,
mated to reproduce her alleged observatibns, as data actually obtained in her
experiments.
        Having considered the evidence fiom the box the subject lefi behind and the
subject's response to the allegation, OIG concludes that the published spectnun, in thesis




  6D.
31
 Letter from the subject to Associate Dean (1 Octokr 1996) (emphasis in original) (exhibit IC).
         figure 3.6(d) (exhibit 6A), figure 3(d) (exhibit 6B), and figure 10(d) (exhibit 6C), was
.        fabricated.

                 Fabricated A n a l y s e s in Thesis khapter Two
                                                             I
                 In a m d y s i s , chemical compo&ds dissolved in solution are separated by
, J -18                             then passed t h r o u i a e c t m m e t e r The different
          compounds pass through the chromatograph kt different rates. The peaks on the
                                 showing the output of )thel o v e r t h e , correspon
         to the different compounds. As each compound emerges h m the
                                                                                                      a
         fed into a mass spectrometer. he -specqmeter             separates
         fragments of the input compound by moleculaf weight. Peaks are obsewed in the mass                .




                                                 -
         spectrum at the mass of each kind of molecule and molecular hgment. These peaks
         identify the compounds dissolved in solution.       I
               Unlike the evidence discussed above c/~ncemingthe-det           erminations and the
      -spectra,      actual cut-and-taped instrument output for the a l y s e s is lacking.
      OIG believes that the evidence provided to us)by the institution, however, supports the
      allegation-and the Committee's conclusion-+at the -analyses                 were fabricated.
     Thesis figure 2.13, an example of the I).analysis             in chapter two of the subject's
     thesis, is attached as exhibit 7A. Two                results, h m the box the subject left
     behind, are attached as exhibits 7B and 7C. Exhibits 7B and 7C were not included in the
     subject's thesis, but are integral to the analysis leading to thesis figure 2.13. The lower
     plots in exhibits 7B and 7C are the mass spectra presented in thesis figures 2.13(a) and
     (c), respectively; the upper plots in exhibits 7B and 7C are                           t were
     not included in the thesis. The faint lines and shaded m&gs          highlighted ontxhibits
     7B and 7C are the "effkct of cutting-and-taping)' described by the subject's PhD. advisor
     a s evidence that th-sults            were fabricated."

               Aside fiom the evidence of cutting-and-taping, the subject's Ph.D. advisor told
      OIG that the experiments leading to thesis f i d e 2.13 were unlikely to produce t h e m
      d      a      t    a in the upper plots of exhibits 7B and 7C or the data in thesis figure
      2.13. First, he pointed out that the                          exhibits 7B and 7C contain
      only single peaks (near 11 minutes) -of                       y s . He told OIG .that if the
      reaction studied in thesis figure 2.13 were analyzed with a                              the
      resulting s h o u l d have included beaks corresponding to other compounds
     during a 25-minute analysis, in addition to th$ single product peaks shown in the #
    i                   n exhibits 7B and 7C. Second, the subject's Ph.D. advisor said that
     another graduate student, who studied the s h e reaction after the subject left the
     institution, found that the reaction actually took huch longer than the 10 minutes claimed
     by the subject in the caption to thesis figure2!13. Finally, the subject's Ph.D. advisor
     explained that e c t r a presented in thesis kgure 2.13 do not reflect the presence of
     expected, naturally o c c u r r i n ~ s ~ t o ~whiche s , would have introduced additional
     peaks (shown as vertical lines) in that figure.

    32
     Letter from the subject's Ph.D. advisor to Chair
          1B).
                                                        01   thi Committee at 2 (25 J k e 1996) (exhibit
             The subject does not contest the chargd that she fabricated th-results."

             OIG's Assessment of the Evidence
                                                         I
             Based on our review of the docume&uy evidence provided by the institution,
      OIG agrees with the Committee that '9,                    and s p e c t r a reported in the
     thesis [have] been altered through cutting and pasting to eliminate act@ data and add
     new spectral features where none appeared Jin authentic spectra."34 The Committee
     concluded that "the heart of [the subject's] dissertation is based on hudulent data" and
     found "a very clear pattern that undermines the entire basis for the research reported in
     the di~sertation."~~

          The evidence overwhelmingly suppoq the conclusion that the subject fabricated
  the spectra and chromatograms discussed          Such fabrications required deliberate
  and carefully planned                                      the objective of misleading
  the subject's institution into                            this is willtul conduct. OIG
  agrees with the Committee that the                 fabrications, which undermined the
  basis for the r e s m h reported in her         the subject to accede to the rescission
  of her Ph.D. degree, are a serious            h m accepted practices and, therefore,
  misconduct in science under NSF's regulation.

         In deciding what actions are appr&riate when misconduct is found, it is
 appropriate for NSF to consider whether &e subject's acts are part of a p a t t ~ 3 7
 Accordingly, we note that several other cut-add-taped spectra ware found in the box the
 subject left behind."      While these cut-anditaped spectra were not included in the
 subject's thesis, they reflect a pattern of fabrication by the subject.
                                                     I
                                 OIG'S RECO&TION
        We recommend that NSF afEm the siriousness of the subject's acts by finding
that the subject committed misconduct in sciehe and issuing a letter of reprimand. We
do not believe fiurher action by the govemmedt is necessary because the actions taken by
the institution are adequate to protect the gov&xnment's interests and the subject advises
us that she has not worked i n m s i n c e l s h e forfeited her degree. We recommend
that NSF develop a notification requirement so that, should the subject perfom federaIly
                                                     I
33
  Letm fmm the subject to OIG (1 September lb98) (exhibit ID); letter from the subject to
                                                    r,
     Associate Dean at 1 (1 October 1996) (exhibit 1, ).
 '4~etterfrom Committee to Associate Dean (25 June 1996) (exhibit 1A).
 35
   Letter fiom Committee to Associate Dean (25 ~ d 1996)   e (exhibit 1A).
MOIG concludes that the subject engaged in miscodduct in science based on her fabrication of the
    determinations, t r a ,                                    as discussed above. The subject does
    not contest the allegations that the                  and t h e data were fabricated, and the
    subject admitted the seriousness of the uncon+sted charges by withdrawing her thesis and
    returning her Ph.D. degree. OIG believes that, even based solely on the uncontested
    allegations, the subject's fabrications constitute hisconduct in science.
"45 C.F.R. 689.2(b)(3).
%ee discussion supra note 8.