When Collaboration Goes Too Far
Whether or not cheating took place in a 100-level course at the University of Pennsylvania's Wharton School last spring, the scenario being described is likely to raise red flags for professors everywhere.
The allegations involve teams of students collaborating on an assignment that required writing complicated computer code, according to students at Wharton, Penn's business school. As first reported Thursday in The Daily Pennsylvanian, the course in question is Operations and Information Management 101, which is required, is made up mostly of freshmen and is known to be one of the most challenging in the school.
Students were asked to work in groups of two or three and not to collaborate outside their teams. But Alex Anderson, a Wharton sophomore who said he took the course and is not being investigated, said he knows of groups that either asked for or shared information with others.
The assignment required that groups show the process they used to reach their answers, and professors noticed similarities, according to Anderson.
"Cases shouldn't be identical because there are a million ways to write code," he said. "It's not expected that groups working on their own would show the same, uncommon way of doing it."
In some instances, one group would send the case -- as the assignment is called -- to another under the assumption that the students wanted ideas but didn't intend to copy directly, according to Anderson. He said he is aware of at least six students who have been asked about their actions, and said that dozens could be questioned by professors in the class. To his knowledge, no students have been charged with cheating.
Jason Toff, a Wharton senior who is a private tutor for the class, said that while he cannot confirm that cheating took place, he's aware that questions have arisen about whether people in different groups coached each other on a section of the case, and whether someone forgot to delete the answer to a case from a public computer.
Penn's Office of Student Conduct, which handles cases of alleged academic misconduct, declined to comment and is not acknowledging that an investigation is taking place. Thomas Lee, an assistant professor at Wharton, said the university makes every effort to protect student confidentiality. "As a general matter, the department and the faculty as a whole take academic conduct very seriously," he said in an e-mail. "If there are incidents of apparent academic misconduct, efforts are taken to respond appropriately including, when appropriate, forwarding details to the [conduct office]."
The Penn allegations come just months after two-high profile cases of cheating surfaced at other universities. Nearly half of the then-second-year students at Indiana University’s School of Dentistry were found to have either taken part in or to have known about and not reported an incident that involved breaking into password-protected files to get an early look at images on an exam. And Duke University’s Fuqua School of Business announced that dozens of first-year students violated the honor code by collaborating on a take-home test that was supposed to be completed alone.
Don McCabe, a professor of management and global business at Rutgers University who tracks cheating, said that data he has collected of thousands of students over the past five years indicate that collaboration is quite common. Roughly 40 percent of the undergraduates surveyed acknowledged working together or sharing information on assignments intended to be completed alone in a way that they say violates a code of conduct. (He didn't ask about group work.)
McCabe assumes that the figure is actually higher, because many probably misinterpret or rationalize their actions as being appropriate. His data show that the majority of students consider working on assignments together when individual work is requested either "trivial" or not cheating.
"I don't know a group that doesn't collaborate in one way or another," Anderson, the Wharton student, said. "Sometimes it's just a push in the right direction. Professors are usually aware of that. There's a difference between that and sending and copying code."
Anderson said that students who turned in the assignment without understanding how their group reached its conclusion were likely to perform poorly on the midterm and final, which require them to replicate the process on a computer. He added that while "no one wants to see classmates get caught," students who didn't cheat might be helped in the end if others perform poorly on tests, since the course is graded on the curve.
Toff said that while the course in question doesn't allow collaboration outside the group, other courses at Wharton encourage learning in larger teams. The Operations and Information Management course offers mentoring and teaching assistant sessions, he said.
McCabe said that what's most telling is how much an assignment counts toward the final grade.
"I've seen collaboration for so many years, particularly if it's on work that's not a big percentage of the grade," he said. "Students don't see it as a big deal. But if it's too much a part of the grade, it's just going to wreck the curve. Groups get protective of their own status."
Toff said the Wharton assignment in question has accounted for 15 percent of the final grade in past years.
David M. Eberhardt, a research associate at Florida State University’s Hardee Center for Leadership and Ethics in Higher Education who writes an ethics column for its Journal of College and Character , said he's fascinated by the extent to which students go or consider going to get assignment answers.
"Students talk about the desire for structure; they want to know what's expected," he said. "I don't know how much more clear a faculty member can be than to say, 'Don't work together.' "