NSF OIG Investigations:
Misconduct, Malfeasance,
and More
Howard University
RCR Workshop
October 2012
Kenneth L. Busch, Ph.D.
Investigative Scientist
kbusch@nsf.gov
703-292-4569
Research Misconduct
Case Study
 University notifies us that data submitted in an





NSF proposal may have been fabricated
The data were from a student conducting survey
research
Results looked very promising — too promising
Mentor submits NSF proposal but then
questions student on veracity of data
Student suggests that proposal be withdrawn
Student claims that data was analyzed by some
unknown individual — data exchanges via email
Research Misconduct
Case Study
 Unknown person sends an email to mentor stating data are made
up, and apologizes. Then the email account is deleted
 The University investigates and determines that the student made
up the data. Student does not defend herself but does not offer up
the identity of the “unknown person”
 Results (after our ROI is sent to NSF)
Student dismissed from University
NSF debars the student for three years
Office of Inspector General (OIG)
 Almost every federal agency/entity has an IG
 An IG is an independent office for oversight
 Promote economy, efficiency, and effectiveness…
 Prevent and detect fraud, waste, and abuse…
…in agency programs and operations
 NSF OIG
 38 audit staff, 22 investigative staff
 Investigations staff includes:
Ph.D. scientists
Special agents
CPAs
Attorneys
 Tasked with investigating research misconduct allegations
Federal Funding Expectations
 NSF/Federal Gov’t
 Clear articulation of rules/expectations
 Balance compliance, institution responsibility and latitude, reduction
of bureaucracy
 Numerous funding opportunities
 Institution
 An environment in which employees can operate with integrity
 Responsible administrative, financial, and research management
and oversight (e.g. Article 1, GC-1, OMB Circulars)
 Principal Investigators




Overall -- Uphold ethics and standards of community
Submit quality proposals and conduct the funded activity
Know and adhere to rules, regulations, and ethics
Ensure compliance and education of staff, students
What does NSF OIG investigate?
The simple answer:
Lying
Cheating
Stealing
How does OIG know
what to investigate?
Allegations received from:







Program officers
Reviewers
Colleagues
Students and post-docs
University administrators
People like you
Anyone with an interest in
what NSF funds
 Anonymous
We take a general look:
 Proactive reviews
 Recurring “problems”
What we don’t investigate . . .
 Generalized QUALITY OF SCIENCE
 Scientific “divorces”
 Patent, copyright or programmatic disputes
 Institutional personnel issues that do not
violate statutes, regulations, or grant
conditions connected with NSF programs
 Authorship disputes
i.e., Whose name goes on the paper? In what order?
Research Misconduct (RM)
 Federal-wide definition and procedural framework.
 RM means fabrication, falsification, or plagiarism in
proposing or performing research [], reviewing
research proposals [] or in reporting research funded
by [the agency]. 45 C.F.R. 689.1.a
Fabrication:
making up data or
results and recording
or reporting them
Falsification:
manipulating materials,
equipment, or processes,
or changing or omitting
data or results
Plagiarism: appropriation
of another person’s ideas,
processes, results or
words without giving
appropriate credit.
 From 1998-2008, NSF has observed a 3-fold
increase in RM allegations
Allegations Since 1998
Fold Increase in Number of Allegations
4.5
*Normalized to 1998
research misconduct
allegations
4
3.5
3
Misconduct In
Science
2.5
2
Research Misconduct
1.5
1
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Year
Research Misconduct
Case Study
 Master’s Student makes up his entire MS research project
 Published 3 articles in major journals with two mentors
 Also published in one conference proceeding
 Successfully completed his MS oral defense
 Slipped up in post-defense celebration
 Claims laziness as an excuse
 Never really interested in the research
 University removes student and voids his MS efforts
 NSF debars him for 5 years
Research Misconduct
Case Study
 Post doc allegedly fabricates data in a plant research project
 Post doc publishes in a major journal

Supplementary data posted online
 Peer researcher reviews research and finds data are
questionable—notifies mentor
 When approached, post doc confesses
 Only one data set existed – the “replicated” two data sets
were simple multiples of original data (.95,1.05)
 Claims pressure to publish and lack of adequate supplies in
laboratory
 Employment terminated, debarred by NSF
Research Misconduct
Case Study
 PI submits proposal w/ copied text and altered figures
 Amount of text copied was relatively small
 Figures of cells were taken from a publication by his former
mentor
 Figure caption was altered to represent three fluorescent
images that did not accurately describe the image
 Text in proposal was written to suggest that images were
preliminary data collected in his lab
 University found research misconduct and PI’s employment
was terminated
 NSF finds research misconduct – letter of reprimand; 3 yrs of
certifications and assurances
Research Misconduct
Case Study
 PI fabricates the existence of two manuscripts in his
biographical sketch and makes reference to a nonexistent manuscript in the text of his proposal
 PI admits that manuscripts did not exist.
 Showed a pattern of deception about these manuscripts in
other scientific proposals submitted to other funding sources
 Institution finds research misconduct and PI’s employment is
terminated
 NSF makes finding of research misconduct – letter of
reprimand and 3 years of certifications
Research Misconduct
Case Study
 Allegation that PhD student fabricated work in his dissertation
 Most work is verifiable……but a small piece is not
 Mentor files complaint with university
 Student asks for obscure information
 Many times refuses to attend hearings regarding the matter
 In his absence, committee finds that he committed RM
 Loses degree 3 yrs after it is conferred
 Loses job as professor at small college
Research Misconduct
Case Study
• PhD student fabricates data in 6 of 16 federally funded
research projects
• Alters data to better support research hypothesis
• Solicits assistance of husband’ to cover up
• During investigation, admits to fabrications and commits
to assist university in cleaning up fabricated data
• Actions/Results
• RM finding, 5 year debarment
• University removes student from PhD program
• Loss of offer to join faculty at Harvard Business School
What was the cost of the University investigation?
Other Concern Areas
 Human subject research
 IRB
 Animal research
 IACUC
 Conflicts of Interest
 Biohazards
 Financial Management
When Administrative cases turn
Civil/Criminal . . .
 PI submitted SBIR-1 proposal as follow up to his
MS Student’s thesis work ($100K, 6 months).
 PI copied the thesis into his final report
and proposal for the SBIR-2 award ($500K).
 University notifies OIG of plagiarism allegation
 When awarded, PI used the money to
pay his child’s tuition at a University, along with
other personal expenses.
 The PI denied everything. His wife did not.
When Administrative cases turn
Civil/Criminal . . .
 NSF suspended the award and OIG issued
subpoenas.
 OIG referred the case to DOJ, which accepted
it for prosecution.
When Administrative cases turn
Civil/Criminal . . .
 At a meeting with DOJ, the professor (through his
attorneys) indicated that he would like to
1) plead guilty to a criminal count (1001)
and pay $240,000 restitution
2) avoid jail
3) avoid Federal action against his wife
 NSF OIG recommended RM finding and debarment.
Professor and NSF settled for 3 years voluntary
exclusion from Federal funding.
Excerpts from the Plagiarism
Hall of Excuses










It’s only background/introductory material (or it had no technical
merit.)
It’s only a proposal. It’s not like it’s a publication.
The reviewers are smart enough to know what is my work and
what is someone else’s.
My English teacher told me it’s not plagiarism if I change every
seventh word.
In my country, this is how we do it.
My native language does not have a word for plagiarism.
I didn’t do it. My grad student/undergraduate/postdoc/grant
writer/faculty colleague/secretary/Co-PI/SRO/AOR/VP of
Research/Dean/spouse wrote that section.
It’s in the public domain.
The pressure of the deadline and the room overheating resulted in
my carelessness with my citations.
It’s not plagiarism; it’s just bad citation.
Your Take-Home Message
 Research community operates under the
assumption that effort is honest and that integrity
standards are upheld
 If you uphold the standards, then you can hold
the next generation to the same standards
 If you lose your professional reputation, it’s hard
to recover
 Federally funded research is a public trust; if not
treated as such, consequences are severe
Don’t lie. Don’t cheat. Don’t steal
Contact Information
nsf.gov/oig
Hotline:1-800-428-2189
E-mail:oig@nsf.gov
Fax:(703) 292-9158
Mail:
4201 Wilson Boulevard
Arlington, VA 22230
ATTN: OIG HOTLINE
Purchase Card Fraud
 Purchase (P) Card misuse at a university,
 Center Accountant - Large volume of purchases




from Internet vendors,
Vague receipts submitted by the P-Card user,
many receipts looked similar,
Questionable purchases charged to Federal
Grants.
Center Accountant resigned after University
Auditors asked about internet purchases.
Investigation initiated by NSF-OIG
Evidence found on the Center
Accountant’s Computer
 E-mails from Internet vendors shipping goods to
the employee’s home,
 Copies of templates for various Internet vendor
invoices,
 One template had been altered 64 times,
 Personal Pictures of the P-Card User enjoying
the good life.
Investigative Findings
 Over 5 year period of time
 Over 3800 personal purchases made from more
than 15 different vendors
 Over 1900 transactions through the institution’s
financial system
 Over 30 different accounts (NSF/State/Private)
fraudulently charged
 Over $316,000 fraudulently diverted
Concealment of Fraudulent
Activities
 Majority of items shipped to home address
 Lied to supervisors and co-workers
 Forged the supervisor’s signature on PCard statement review documents
 Created false invoices
 Manipulation of the institution’s accounting
system
How many items were
purchased with Grant Funds?
By tracking the evidence, we found
20 items purchased with Grant funds.
What was really bought
DOJ Resolution
 22 count Federal Criminal Indictment
 17 Counts “Mail Fraud”
 5 Counts “Theft from an Organization Receiving
Federal Funds”
 Guilty Plea to all 22 counts
 Sentenced to:





32 months prison
3 years supervised probation
$316,000 Restitution/$2200 fines-assessment
250 hours community service
5 year debarment.