The notion that students should collaborate with one another outside of class has been embraced nearly universally in higher education, but is this a good thing? And when is it most appropriate?
Those are some of the questions prompted by the news last week that Harvard University is investigating 125 students for allegedly cheating on a take-home final examination last spring.
The allegations of academic dishonesty on the open-book, open-note, open-Internet exam range from “inappropriate collaboration to outright plagiarism,” Jay M. Harris, dean of undergraduate education, wrote in a letter to students. Nearly half of the more than 250 students in the course “may have worked together in groups of varying size to develop and/or share answers.”
The university said the incident “betrays the trust upon which intellectual inquiry at Harvard depends.” In a statement to The Chronicle, Harvard said that a committee of faculty members and deans was reviewing the allegations and would meet with each student under investigation over the next several weeks. Some students may be asked to withdraw, depending on the offense.
“We expect to learn more about the way the course was organized and how work was approached in class and on the take-home final,” the university said. “It will take time and require patience, and we cannot comment on specific cases.”
Harvard has not specified the class under investigation. The Harvard Crimson identified it as a government course, “Introduction to Congress,” which is taught by Matthew B. Platt, an assistant professor.
While Mr. Platt expressly forbade his students to discuss the final exam with others, some students said they were routinely encouraged to collaborate throughout the semester.
“My understanding was that you shouldn’t sit down and take a test with someone else, or take someone else’s test and present it as your own,” one student told The Boston Globe. The student said that despite the ban on discussing the exam, he figured it would still be safe to share a few ideas with other students. “But I wrote my own answers on the final.”
Such confusion is common, said Donald L. McCabe, a professor of management and global business at Rutgers University in Newark and New Brunswick, N.J., who has studied student integrity.
“There is a gray area there,” he said. Faculty members often extol the merits of group work, but don’t always distinguish between permitted and impermissible collaboration. “My view is they have to be explicit in their instructions on the assignment.”
An editorial in the Crimson also pointed out that ambiguity but didn’t let students off the hook. “The definition of collaboration has become increasingly nebulous, and the difference between acceptable and unacceptable collaboration even more so,” the editors wrote. “Take-home exams in particular have become uncomfortable gray areas in which both spheres often overlap.”
And yet, they noted, the Harvard students in this case had clear instructions from Mr. Platt, which should have overridden any sense that they had implicit permission to collaborate. “If these transgressions did in fact happen,” the editorial said, “they were not the product of a misunderstanding, but the willful defiance of course policy.”
‘Implementation Matters’
Undergraduates have widely embraced working in groups as a way to learn. It is one of several practices that indicate student engagement and is measured in the National Survey of Student Engagement, or Nessie. It is thought to shift the educational focus to students and to train them to work in groups, a practice they will often encounter once they embark on their professional lives.
Nine out of 10 students work with other students outside class on assignments, and more than 60 percent do so “often” or “very often,” according to Nessie data.
In practice, however, working in groups has come to mean many different things that may go by the same name but are not of equal value, said Alexander C. McCormick, director of Nessie and an associate professor at Indiana University at Bloomington’s School of Education.
“Group work is a good thing, but like almost anything that’s good, the ‘how’ is at least as important as the ‘what,’” he said. “In other words, implementation matters.”
A well-structured, clearly defined group project, in which students frequently check in with their professor, typically carries significant educational value, he said. Studying with friends and working on weekly problem sets together may not do very much. Existing Nessie data do not distinguish between those two kinds of group work, though forthcoming surveys ask for more detail from students.
While it is good that colleges are paying closer attention to what Nessie identifies as the “high-impact practices” that are thought to carry educational benefits, Mr. McCormick said he worried that some institutions resorted to a “checklist mentality” in which they adopted practices without making them part of a coherent strategy.
“Some surely do it well and carefully and thoughtfully and intentionally,” he said, “and others may not.”
Some evidence suggests that more time spent studying with peers actually predicts drops in student learning, particularly if it is not part of a highly structured and thoroughly conceived assignment, said Richard Arum, a professor of sociology and education at New York University and an author of the book Academically Adrift: Limited Learning on College Campuses.
For the book, Mr. Arum and his co-author, Josipa Roksa, an associate professor of sociology and education at the University of Virginia, studied the effect that several variables had on a test of critical thinking, the Collegiate Learning Assessment.
They found that the more time students spent studying with peers, the more their scores declined, as compared with other activities, such as working on or off the campus, or participating in student clubs. If they studied with peers for 20 hours in a week, their scores dropped by as many as 80 points over two years.
“We found nothing else that was remotely close to that,” Mr. Arum said in an interview on Wednesday. “That’s more than the entire growth we saw for the sample as a whole.”
For Mr. Arum, the most troubling aspect of the allegations at Harvard is the low academic expectations that some students perceived for the course. One student said that people seldom attended lectures or discussion sections, but many could still expect to receive A’s.
“There was absolutely no incentive to learn any material,” one student told Salon.
“Here we see at Harvard,” Mr. Arum said, “at arguably one of the best colleges in the country, very disturbing evidence of what academic life looks like.”
