Every time Rodney* ’19 received an essay assignment in his sophomore English class, he would go home and share the prompt with his tutor, who would write the essay in front of him on his own computer while only sometimes asking Rodney for his input.
“I got worse [in English last year] because I stopped having to care or try,” Rodney said. “I would know every day that the in-class discussions were absolutely meaningless for me because what we discovered in class would have nothing to do with what I would write about in my essay. I wouldn’t even know what I’d write about.”
Although Rodney’s essays were not his own work, all of them were technically original, so he has never been caught cheating by the school’s plagiarism detector, Turnitin.
Penelope* was not so lucky. She was brought to the Honor Board last year after her English teacher found that she plagiarized her analysis almost directly from a website. While she said many teachers can spot plagiarism based on writing style, she thinks her teacher knew she was cheating largely due to Turnitin.
“The software is very good,” Penelope said. “Even if it’s not verbatim, it could be synonyms, or even if it’s not in the exact same order, the algorithm they use is still clever enough to detect plagiarism.”
Even though she no longer plagiarizes on essays, Penelope said she worries that the algorithm could generate false positives, resulting in further accusations.
In a Chronicle poll of 314 students, 76 percent of respondents said they think using Turnitin is an effective way to detect cheating.
But some students have discovered loopholes within the Turnitin system. Guillermo,* who has never been caught cheating on an English essay, created a Turnitin teacher account during his sophomore year to test all of his essays for a plagiarism notification. To keep students from testing their assignments, Turnitin does not let the general public create these accounts. However, after his sophomore year, Guillermo looked through the faculty directory at a high school in Texas and posed as a random teacher to get his own account and submits all of his essays through fake Turnitin assignments to check their amount of original content.
“You always want to take every precaution,” Guillermo said. “Some people are like, ‘I’m kind of worried about Turnitin,’ and I say there’s no need to be worried. Before any assignment, you should already have everything under control. I just thought it would be a good idea to make sure I wouldn’t get caught.”
He said he often consults online sources before writing essays but is careful when integrating them to change enough not to get caught.
“It’s not paraphrasing in that I copy and paste it and change a few words, I really try to rewrite it in my own way and Turnitin can’t catch that,” Guillermo said. “All Turnitin does is do a search of online stuff. Of course it says it’s really sophisticated, but there’s a point where it can’t know.”
Essays are not the only assignments that Guillermo cheats on; he has also employed his cheating tactics when completing science labs. Using a fake email address and refusing to communicate over text or email to make sure there is no paper trail, he has older students send him their finished assignments from courses they have already taken. Guillermo said he is confident that he reworks phrasing enough to ensure that he does not get caught.
For Duncan,* not cheating on labs is less common than copying someone else’s work. He estimates that he has done 10 percent of his labs at the Upper School alone. Even so, he has never been called to the Honor Board or questioned by a teacher.
“[Students] go to ridiculous lengths not to be caught,” Duncan said. “Changing the bulletpoint format, so that when one person has dots, you have dashes, using Ctrl+F to replace certain words with different words with different punctuation, anything to make your lab slightly grammatically different from another person’s.”
Duncan and Guillermo are far from alone. In a Chronicle poll of 315 students, 36 percent of students say they regularly consult finished labs, and eight percent say they consult finished essays before turning their own assignments in.
Earlier this year, after Jonathan* tried to help a younger friend with a science lab who could not finish it, he sent a copy of his lab to the friend. The friend was caught after turning in the entire lab, including questions that were not included on this year’s lab.
“A lot of times, there’s a gap in what people think is actually wrong for labs,” Jonathan said. “There’s a lot of really clear, blatant cheating, but then also sometimes people get confused and can accidentally cheat.”
Teachers are well-aware of the cheating epidemic. Physics teacher Karen Hutchison said she has had several instances of academic dishonesty on labs this year. Hutchison estimates that 25 percent of AP Physics I labs are done within strict Honor Code boundaries and that 50 to 60 percent of her students have not consulted completed labs that are not their own before turning reports in.
As a result, she said she has considered alternative lab exercises to stem dishonesty on labs.
“We could, for example, require all of the work to be completed in class during that period and turned in at the end of the period so that there’s no opportunity or less opportunity for that type of sharing,” Hutchison said. “We could also just have students turn in a single lab report for a lab group so that that type of sharing is rendered unnecessary.”
Alternative lab exercises would mean less in depth lab reports or more time spent in class on lab assignments, Hutchison said.
She also said she is tentatively considering having students turn all AP Physics I lab reports in through Turnitin next year.
“[Hopefully] we can try to catch some of the similarities between lab reports from that year and maybe try to emphasize to students that this is something we take seriously,” Hutchison said.
Science department chair and AP Biology teacher Larry Axelrod includes a “Plagiarism Statement” on his class hub page.
“Discussion for the purpose of understanding elements of a problem is fine, but it is not a one-sided conversation that supplies you with an answer,” the document states. “Furthermore, discussion precedes answering the question; it does not coincide with it. When you physically sit down to write your response, you do it alone.”
For now, some members of the science department are in the discussion stage about how to evaluate student understanding of lab material, but no official policy has been finalized.
*Names have been changed