A student’s programming code just won’t run. It’s getting late, and the assignment is due in just a few hours. There are a million other things to do. The specter of a failing grade looms large. And lifting part of a classmate’s work before clicking submit seems like an easy shortcut.
Taking such a step — and getting caught — could result in a disciplinary hearing and a harsh sanction, but not necessarily in Harvard University’s wildly popular introductory computer-science course. Professor David J. Malan has incorporated a “regret clause” into his syllabus: If first-time offenders come forward and admit what they did within 72 hours, an instructor will give a failing grade on the assignment — but will not refer the case for disciplinary action.
Six years in, the clause — used by a tiny minority of students — has not pushed down the percentage of students in the class referred to the university’s honor council, according to a paper Malan released recently. But he has learned some valuable lessons about why students cheat, and he believes conversations with regretful students may lead them to develop healthier work habits, like reaching out for help or attending office hours. He recommends that other instructors, even outside computer science, adopt the initiative for that reason.
“Acts of academic dishonesty were a symptom of larger concerns or pressures in their life,” Malan said of some cases. The conversations, he said, “made it much more real, and much more difficult, because now you are on the front lines, discussing these things with students.” Sometimes, in the conversations, the student cries.
The paper, released in December, offered a comprehensive set of statistics on the policy’s use and wider effects for the first time. Hundreds of students enroll in Malan’s course — CS50 — each fall, and the number of students who invoke the regret clause annually peaked in 2015 at 26, or about 3 percent of that fall’s 750-person class. This past semester, just 8 of 781 students did so.
Malan expected the number of cases that course leaders would pass along to the honor council to decline, but it has actually increased, which he believes is because he feels more comfortable referring students to the council. After all, he had already given students an out to admit their wrongdoing.
The practice has expanded to other campuses, including the University of Utah, the University of Massachusetts at Amherst, and, recently, the University of Texas at Tyler. Robert P. Schumaker, a professor there, saw Malan’s paper in December and decided this semester to incorporate the clause into his computer-science classes. Toward the end of the semester, he said, students’ “common sense goes out the window” as they work on projects late at night.
The regret clause, in addition to other language in his syllabus — including guidelines from Malan’s Harvard class about what type of collaboration is reasonable or unreasonable — clear up ambiguities for students, Schumaker said.
Despite Malan’s experience at Harvard, Schumaker said he hopes the guidance will reduce the number of judicial-affairs cases he has to submit. “I go through that every semester — it is tough,” he said. “I don’t want to have those situations in my class.”
Shriram Krishnamurthi, a Brown University computer-science professor, has never had a student invoke the clause, which he in part attributes to the fact that he is stern about academic honesty at the outset of his course. Having a “regret clause” on his syllabus shows a more compassionate message: that he understands people are human.
I’d much rather spend my time on the students who need help than 10 times that time documenting plagiarizing cases.
“I’ve also come to realize I’d much rather spend my time on the students who need help than 10 times that time documenting plagiarizing cases,” he wrote in an email. Of course, if he finds evidence of academic dishonesty, he pursues cases in earnest, but he wrote that “many faculty feel a certain degree of fatigue and pointlessness to reporting, and remind themselves they took up this job to teach and grow students, not to join a police department.”
Creating a disciplinary process outside the university’s central procedures raised some concern initially at Harvard. One big reason for that is that a large number of cheating-case referrals at the university come from computer science. From 2015 to 2019, Harvard’s school of engineering and applied sciences referred more cases to the council than did the humanities and social sciences combined.
Before rolling out the policy, Malan said he discussed it with administrators and the council itself, a “delicate” process. He felt he needed to ensure those offices were comfortable with the language. After those conversations, the course’s leaders made clear that the clause would apply to first-time offenders only.
That tension — between universitywide policy and more individualized approaches — exists for many professors, said Susan D. Blum, an anthropology professor at the University of Notre Dame and the author of My Word! Plagiarism and College Culture.
Many professors, swamped with their own workloads, may be more likely to look the other way instead of undertaking a time-consuming disciplinary process. “We don’t tend to get any incentive for spending a lot of time dealing with student misconduct,” she said. “Is it worth my time going through the process? Almost certainly not.”
Therefore, she said, a conversation about at least a handful of these cases a year may help students correct course — a far more likely outcome than if a professor just ignores the behavior because it’s too much trouble.
In 2012, after 125 students were suspected of cheating on a Harvard take-home exam, Blum wrote that students at highly selective universities have a sense of “inevitable achievement” and perhaps feel “entitled to succeed.” She said recently that such an atmosphere at selective institutions could make statements like the “regret clause” more challenging to carry out, though she said that this outlook is far from universal and that she had grown more sympathetic to students since then.
Over all, Blum said, the Harvard “regret clause” merits wider consideration: “They really are trying to get to the bottom of what’s motivating the behavior. Is it that students really need some mental-health counseling, or do they need to improve their academic skills? That seems to me humane.”
