College sophomore and pre-medical student Poojita Chinmay spent weeks studying for her first chemistry exam this September. When she got her grade back, she was alarmed to see that the class average on the exam was 74%. According to her professor, it was a 20-point bump from the 50% average in years past.
“The cheating is pretty obvious," Chinmay said. "It’s not like virtual learning is suddenly so much more effective than in-person learning. As someone who didn't cheat on the exams, I'm just concerned that I'm falling behind, because I'm taking the moral route."
CHEM 102: General Chemistry II, a core course requirement for students on the pre-med track, uses a curved grading system which determines students' grades based on their exam performance relative to that of the rest of the class. Chinmay said that the stress of acing a pre-med course, coupled with the pressure of having to compete against peers for a high exam score, creates a perfect storm that motivates students to cheat and leaves their rule-abiding classmates at a significant disadvantage.
“Some of these kids will do anything to get the grades that they want," Chinmay said. "These are classes that I feel like I would perform well in person for, but now I'm even having to consider if I should pass/fail this class because of the cheating."
Chinmay’s experience is just one example of what many students allege to be a pervasive increase of cheating among Penn students ever since the pandemic forced classes online. The Office of Student Conduct similarly reported a 72% increase in cheating case investigations from the 2018-2019 to 2019-2020 academic year.
In past years, when exams have been administered in person, cheating was much easier to spot and much harder to pull off. Gone are the days of a student’s eyes looking astray at his peer’s exam or students bringing notes into the testing room. Virtual exams are an entirely different ballgame.
How professors are crafting their own anti-cheating measures
Lacking clear policies or guidance from the University, some professors have resorted to implementing their own anti-cheating measures, with varying degrees of success. These measures range from "live proctoring," open-book policies which allow students access to their notes, to lockdown browsers which lock students onto the Canvas browser and prevent access to other online resources.
But without clear evidence of students cheating, professors are left to interpret higher class exam averages on their own, with some attributing the higher scores to other factors like open-book exams.
Chemistry professor Ivan Dmochowski, who teaches CHEM 101: General Chemistry I, contrasted anti-cheating measures this semester with those from last semester, where professors had to abruptly adapt their courses to an online format and, consequently, did not have time to counter potential widespread cheating. This semester, Dmochowski chose to “live proctor” his exams and make them open-book after receiving reports of cheating in students’ end-of-semester evaluations last spring.
“I think a lot of that [cheating reported last semester] was because people were given multiple hours in which to take their exams," he said. "It increases the temptations to check your answers with a friend before submitting it."
Dmochowski administers his 90-minute, open-book exams over Zoom, where students are encouraged to keep their cameras on and can ask him questions during the exam. While he does not require students to keep their cameras on, citing that not all students have access to web cameras, Dmochowski believes that mandating students be present on Zoom, in conjunction with the strict time limit and open-note format of the exam, prevented cheating. About 95% of the class still chose to keep their cameras on during the duration of the exam.
He holds multiple sittings of the exam to accommodate students in different time zones, with different versions of the exam for those taking it outside the normal class time. Anticipating that students would take advantage of the Internet and supplementary materials, Dmochowski decided to make his exams open-book this semester.
“I think you have to design exams expecting people to use their notes or their books or the Internet, because if you don't, then students might cheat or have an unfair advantage," he said. "In many ways, it ends up being a more real world situation, because, of course, in the real world, we have access to the Internet.”
Dmochowski's open-book exams administered this semester feature more open-ended questions that may require students to look up certain information online, like the density of ethanol, which would have been given to students on a closed-book exam offered in-person.
Similarly, History professor Brent Cebul, who teaches HIST 153: The Transformation of Urban America, administered a 90-minute, open-book exam in one sitting over Zoom during his synchronous class time. Cebul said that students who referred to their notes or readings would struggle to finish the exam due to the strict time limit, and he encouraged students to prepare as they would for an in-person, closed-note exam.
Because his exams are comprised mostly of free-response questions, Cebul believes cheating on his exams would likely be more difficult than on STEM exams.
“Students are still having to write their own answers, come up with their own arguments about change over time," Cebul said. "I think on some level, there's a degree of insulation in our particular humanistic field, versus something in the harder sciences or math."
Cebul added that one of the primary benefits of administering his exam over Zoom was that students could ask him questions about the free-response prompts to which he could then answer for the entire class at once.
Although CHEM 101 is traditionally graded on a curve, Dmochowski, as well as Cebul, are not using a curved grading system, which they worry could create a “cutthroat” and competitive learning environment.
A rise in cheating for the 2019-2020 year
The Office of Student Conduct reported in its FY 2020 Annual Disciplinary Report that the number of academic integrity violations in the 2019-2020 academic year, which include plagiarism, cheating, and submission of false data, increased by 17% from the previous year.
Case investigations of cheating saw a 72% spike from 52 cases in the 2018-2019 academic year to 89 in the 2019-2020 academic year when classes moved online in the spring. Similarly, case investigations of "unauthorized collaboration/use of another person’s work" nearly doubled from 34 to 63 in the past year. Because exams are taken virtually, however, the number of actual cheating incidents is likely much higher.
The University Honor Council President and College senior Sarah Simon said, however, that the UHC has not seen an increase in hearings this fall semester. The UHC consists of 25 undergraduate students who work closely with the OSC and sends representatives to sit on student conduct hearings.
“The fact that academic integrity is a deeply ingrained value within the Penn community, in conjunction with the professors’ measures to accommodate these new circumstances, means that cheating is likely not any more prevalent than it is in normal years," Simon said.
Yet many students paint a different picture.
‘Naivete’: Students allege professors are ignorant of cheating this semester
Many students claim that the anti-cheating measures implemented by professors like Dmochowski and Cebul are the exception to the norm, and that many professors have failed to take necessary precautions against cheating even in the face of inflated class scores.
College and Wharton sophomore Carson Sheumaker said he estimates that at least half of the students in STAT 102: Introductory Business Statistics cheated on the first exam this semester. Because students had 13 hours to complete the 25-question exam — of which there was only one version, he believes students took advantage of the long time window and single version exam to collaborate with peers.
"I heard so much about people reaching out to other students and comparing answers," Sheumaker added.
For College and Wharton sophomore Ananya Dewan, the unusually high class average for the first BIOL 221: Molecular Biology and Genetics exam was likely a result of both the exam's open-book format and student collaboration through the form of texting and video chatting during the exam.
“I'm sure that there probably are some people who are cheating, just because the professor mentioned in previous classes that the average hasn't been above 75%, and our last midterm’s average was about 85.5%," she said.
Although students were required to take the exam during the scheduled class time, the exam was not proctored over Zoom, which Dewan believes could have prevented much of the cheating that occurred through peer collaboration.
She added that the exam was graded on a curve, putting those who did not cheat at a disadvantage.
Most students agreed that curved grading is at the crux of the cheating problem. Simply knowing that peers are cheating is incentive enough to cheat in a curved class, as cheaters' inflated scores put everyone else at a disadvantage, College sophomore Sophia Tran said.
“In curved classes where your grade is partially dependent on how well you do amongst your classmates, having an unfair advantage definitely hurts your peers," Tran said. "If it wasn't curved, then you will be benefiting yourself, but cheating in curved classes is at the detriment of others.”
Similarly, College sophomore Kruti Desai said that the class average on the first PHYS 151: Principles of Physics II: Electromagnetism and Radiation exam was an 87%, which former PHYS 150/151 students described as an abnormally high score compared to that of previous semesters.
Because the professor gave students two hours to complete an exam that was designed to take 40 minutes to complete, Desai suspects that students had ample time to call and collaborate with one another.
Although Desai believes the primary way students are cheating is through collaboration, paid homework helper services such as Chegg – which costs users $14.95 per month to submit questions and receive answers from tutors – benefit those with the luxury to pay for these cheating tools.
“I’ve noticed that as the day progresses towards the end of the 24-hour window [in which we can take the exam], the exam questions will start popping up on services like Chegg," Desai said.
Desai called on professors to wake up and see what she and many other students believe is obvious wide-spread cheating.
“I’m surprised that professors don't realize cheating is happening from the anomalies in the averages," she said. "I mean, you have exam averages that are like 20 points higher than what they've been in the class for maybe six or seven years. Why would you just ignore that?”
Inside the struggle to find effective anti-cheating measures
While many students praised “live proctoring,” during which exams are administered live over Zoom, some students like College sophomore Rebecca Hennessey view it as an invasion of privacy.
“[Live proctoring over Zoom] would give me so much anxiety," she said. "I feel like that would just add an increased pressure that I'm not allowed to look a certain way, because they're gonna think I'm looking at something and cheating even though I just need a break from my screen for 10 seconds."
Almost no anti-cheating measure is infallible, though some measures can be thwarted more easily than others, some students said. The use of “lockdown browsers” that prevent test-takers from Googling answers on their computer can easily be bypassed if students have a secondary device to look up answers online, Sheumaker said.
Some anti-cheating measures have the potential to put rule-abiding students at a disadvantage.
For exams in ACCT 101: Accounting and Financial Reporting, students are not allowed to move backwards after completing a question on Canvas, Sheumaker added. While the tactic hinders collaboration due to the randomized question order, it also prevents students from double-checking their work, he said.
College sophomore Heather Schneps believes that professors will be hard-pressed to find an anti-cheating solution that gets at the root of the problem: Penn's competitive culture.
"Cheating is partially a cultural issue," she said. "At a school like Penn, everyone is used to doing well, and the fear of not continuing on that trend can take over for some people."
Regardless of which anti-cheating measures professors choose to implement, Chinmay, like many other students, said that professors cannot remain ignorant of the cheating that has become so ubiquitous since classes moved online.
“It's just naive to trust that students have the right intentions, especially when so much pressure is being placed on them during this time," she said.
All comments eligible for publication in Daily Pennsylvanian, Inc. publications.