Brilliant or Plagiarized? Colleges Use Sites to Expose Cheaters
-Also see: Lessons in Internet Plagiarism, June 28, 2001.
By VERNE G. KOPYTOFF
DR. DAVID PRESTI, a professor
of neurobiology at the University of California at Berkeley, had
always assumed that some
of his students turned in term papers copied from the Internet. After all,
there
are dozens of Web sites
where students who have no qualms about plagiarism can download
ready-made term papers about
topics like ''Hamlet'' and the Russian Revolution to use as their own.
Until recently, however,
Dr. Presti had never caught a cheater because it would have required spending
countless hours searching
the Internet for evidence. But now, using Plagiarism.org, he can identify
students who choose deceit
over research.
Through Plagiarism.org, Dr.
Presti found that 45 of 320 students from the last spring semester had
plagiarized at least part
of their essays from the Internet. ''It's so easy to cheat now,'' Dr. Presti
said.
''But this increasing digitalization
is also making it easier to find cheaters out.''
A handful of companies, like
Plagiarism .org, are offering Internet-based antiplagiarism technology
that
teachers can use for a fee.
The most complex sites compare student term papers with millions of Web
pages and the archives of
dozens of online sites that offer term papers free. While the anti-plagiarism
sites do not have access
to the databases of operations that sell term papers, they would be able
to spot
many commercial term papers
because they can check databases that professors are starting to use to
keep copies of term papers
from past semesters.
If the service finds similarities,
it notifies the teacher, who must then decide whether the similarities
are
coincidences, justified
by proper footnotes or outright dishonesty.
Companies that sell anti-plagiarism
services say dozens of schools are testing such services. Fees start
at about $20 a year for
a class of 30, with cheaper per-student or per-class rates for larger contracts.
The fees are paid by universities
or teachers.
Some students criticize the
technology, saying it undermines honor codes based on trust between
students and faculty.
Plagiarism.org was developed
by John Barrie, a graduate student in biophysics at the University of
California at Berkeley.
He said he had developed the site to ''level the playing field for honest
students.''
Papers sent to Plagiarism.org
are checked by a computer, which looks for phrases matching those from
other sources or are partly
altered (www.plagiarism .org). The computer compares the term papers with
the archives of free online
cheating sites. The computer also does Web searches to look for similarities.
It also compares essays
with papers from previous semesters and other universities.
Within 24 hours, the company
sends a report of its findings to the teacher by e-mail. Teachers are
cautioned by the companies
not to use that information as absolute proof of plagiarism. The reports
merely point out phrases
that should be examined more closely. Teachers must check for themselves
whether flagged sentences
are attributed.
Some cheaters may try to
evade detection by stealing only a few paragraphs, changing words or
inserting the plagiarized
material into the middle of a term paper. But Matt Hunter, founder of the
Essay
Verification Engine, or
EVE (www.canexus.com), an antiplagiarism service based in Sackville, New
Brunswick, said that software
like his usually uncovered even subtle dishonesty.
''My software takes an essay,
fragments it, and if a student has changed the words, it still finds the
pieces,'' said Mr. Hunter,
who is a college student.
This is not the first time
that technology has been used to uncover cheating. A system that has been
in use
for years, the Glatt Plagiarism
Screening Program, lets a teacher find out if a student is truly the author
of a paper by posing a test:
parts of the paper are deleted at random by the program, and the student
is
asked to replace the deleted
material (www.plagiarism .com). If the student cannot do that, plagiarism
is
suspected.
Another program, called MOSS
(www.cs.berkeley.edu/aiken/moss.html), which has been around since
1994 and is in widespread
use, serves computer science classes. It automatically searches lines of
computer code for similarities.
Currently, many schools do
little to stop plagiarism, said Jeanne M. Wilson, president of the Center
for
Academic Integrity, whose
member institutions include Duke University, the University of California
at
Los Angeles, the University
of Maryland at College Park and Rutgers University. Ms. Wilson is also
director of student judicial
affairs for the University of California at Davis. Ms. Wilson said that
some
students had threatened
lawsuits over the issue of plagiarism charges but that she knew of no lawsuits
that had been filed.
''If you look at academic
integrity problems at many campuses,'' Ms. Wilson said, ''there aren't
that many
cases being examined, even
though we think there is a good amount of cheating going on.''
Anti-cheating technology
is not embraced by everyone. At Stanford University, the student newspaper
recently said that Plagiarism.org,
which administrators are considering using, would violate the school's
policy against using ''unusual
and unreasonable precautions'' against cheating.
''The honor code is one of
the most important tenets of the university,'' said Gil Lopez, the newspaper's
editor in chief, in an interview.
''Using technology to catch people contradicts the spirit of the
university.'' Mr. Lopez
said he also worried about the reliability of antiplagiarism technology.
The antiplagiarism companies
agree that technology has limits. Warren Brantner, president of
IntegriGuard, an anti-cheating
service in Harrisburg, Pa., said that it was best to use antiplagiarism
Web
sites with other measures,
like assigning more unusual topics (www.integriguard.com).
Dr. Presti, the neurobiology
professor at Berkeley, said that nearly 15 percent of his students had
plagiarized even after he
warned them about computerized scrutiny.
''They ranged from sloppy
citations to, in one case, the entire paper was taken from several Web
sites,''
Dr. Presti said.
''I went back and talked
to the most serious violators, and they immediately admitted to it,'' Dr.
Presti
said. ''When you show the
color coding and where it came from, there's no denying it.''
An example from Rutgers Camden. A student submitted this assignment:
2. Write a paragraph explaining Why
Determinism Cannot Model Complex Systems, based on ideas from
What is Chaos. Type your paragraph
here:
Chaos is qualitative in that it seeks to know the general
character of a systems long-term behavior, rather than
seeking numerical predictions about a future state. What
characteristics will all solutions of a system exhibit? How
does this system change from exhibiting one behavior to
another?
Chaotic systems are unstable since they tend not to resist
any outside disturbances but instead react in significant
ways. In other words, they do not shrug off external
influences but are partly navigated by them.
The variables describing the state of a system do not
demonstrate a regular repetition of values and are therefore
aperiodic. This unstable aperiodic behavior is highly
complex since it never repeats and continues to show the
effects of the disturbance(s).
These systems are deterministic because they are made up
of few, simple differential equations, and make no
references to implicit chance mechanisms. This is not to be
completely equated with the metaphysical or philosophical
idea of determinism (that human choices could be
predetermined as well).
Finally, a dynamic system is a simplified model for the
time-varying behavior of an actual system. These systems
are described using differential equations specifying the
rates of change for each variable.
3. Define the following
I typed "chaos is qualitative in that it seeks" into Google, getting
the following:
The following words
are very common and were not included in your search: is in that it. [details]
Searched the web for "chaos is qualitative
in that it seeks".
Results 1 - 5 of 5. Search took 0.40 seconds.
Chaos
... Chaos is qualitative in that it seeks
to know the general character of a system's
long-term behavior, rather than seeking
numerical predictions about a future ...
www.exploratorium.edu/complexity/lexicon/chaos.html
- 4k - Cached - Similar pages
The link to the exploratorium had gone
dead, but Google had the file in its cache:
We borrow a working definition for
chaos theory from Dr.
Stephen Kellert: The Qualitative
Study of Unstable Aperiodic Behavior
in DeterministicNonlinear dynamical
systems.11 I should briefly
dissect some of these terms to better
describe what is and what is
not chaotic in nature:
Chaos is qualitative in that it seeks to know the general
character of a system's long-term behavior, rather than
seeking numerical predictions about a future state. What
characteristics will all solutions of a system exhibit? How
does this system change from exhibiting one behavior to
another?
Chaotic systems are unstable since they tend not to resist
any outside disturbances but instead react in significant
ways. In other words, they do not shrug off external
influences but are partly navigated by them.
The variables describing the state of a system do not
demonstrate a regular repetition of values and are therefore
aperiodic. This unstable aperiodic behavior is highly
complex since it never repeats and continues to show the
effects of the disturbance(s).
These systems are deterministic because they are made up
of few, simple differential equations, and make no
references to implicit chance mechanisms. This is not to be
completely equated with the metaphysical or philosophical
idea of determinism (that human choices could be
predetermined as well).
Finally, a dynamic system is a simplified model for the
time-varying behavior of an actual system. These systems
are described using differential equations specifying the
rates of change for each variable.
-------------------------------------------
New York Times Story, September 3, 2001:
MEDIA TALK
An Accusation of Online Plagiarism
By FELICITY BARRINGER
Ken Layne, a columnist with The
Online Journalism Review, at the
University of
Southern California's
Annenberg School
for Communication,
went to the
Web last week to shame a
Business Week
Online reporter who,
he believed,
had stolen his work
without credit.
As plagiarism
accusations go, Mr. Layne's article, "Want My Story? Help
Yourself!" had
more heft than some. Comparing his Aug. 14 article about an
Australian Web
site and an Aug. 28 Business Week Online article that mentioned
the site, he
found suspicious similarities.
Both pieces used
identical language in reporting that the founder of the Web site,
Crikey.com.au,
was awarded a prize for 16 articles detailing "his adventures as a
loudmouth shareholder
in 50 of the country's biggest companies." Each article
compared Crikey.com's
costs to those of salon.com, saying that salon loses $30,000
a day.
But nowhere in
Ms. Black's article was there a sentence exactly identical to any of
Mr. Layne's.
Ms. Black was out of the country late last week and did not return a
call and an
e- mail message seeking comment. Bob Arnold, the editor of Business
Week Online
said Mr. Layne "is exaggerating." He added, "There are a couple of
parallels that
I wish weren't there. But it doesn't amount to plagiarism."
In his formal
response, posted at www.ojr.org, Mr.
Arnold wrote,
"publishing independently verified
facts that previously
appeared elsewhere isn't
plagiarism."
Since the dispute,
Business Week Online has credited
Mr. Layne's
work.
Want
My Story? Help Yourself! On the Internet no one can hear you steal
By Ken Layne, OJR Columnist