I teach philosophy to college students, and there was no way I was going to give them exams this semester, with our classes being held online. Why not? Simple — cheating. It is nothing personal with these particular students, but I have read enough psychological research to know that it would be very hard for them to resist looking for help in places where they are not supposed to, such as their notes, their friends and the internet.

I am fortunate that papers are a great alternative means of assessment in philosophy courses. But they do not work so well in certain other fields, like the sciences. In this time of widespread online learning and home-schooling, what can be done to curb cheating on exams?

One solution is remote proctoring, where the student is video-recorded during the exam, with any suspicious web browsing reported. That might be effective, but it strikes me as a crude approach, relying as it does on active surveillance, which creates an overt atmosphere of distrust. Naturally enough there are also privacy concerns, as well as some anecdotal evidence that remote proctoring technology encodes racial biases.

Instead I suggest that a practice that has been used widely in other educational contexts be extended to the world of online testing: pledging one’s honor. Honor pledges not only are surprisingly effective in curbing cheating; they also promote honesty. Students who abide by them refrain from cheating not because they can’t, but because they choose not to.

It is easy to be cynical about honor pledges and honor codes. They can seem to be — and sadly too often are — P.R. stunts for schools looking to burnish their image. Or administrative mandates that do not have buy-in from the faculty. Or just a formality, where students check a box on a form during first-year orientation and then never give it any thought for the rest of the year. Honor codes like these are indeed mere facades.

But many schools and programs, from elementary to graduate level, take their honor codes seriously. And for good reason. Empirical research has repeatedly found that schools that are committed to honor codes have significantly reduced cheating rates compared with schools that are not.

Donald McCabe at Rutgers Business School and Linda Treviño at the Smeal College of Business at Penn State found a 23 percent rate of helping someone with answers on a test at colleges without an honor code, versus only 11 percent at schools with an honor code. They reported impressive differences as well for plagiarism (20 percent versus 10 percent), unauthorized crib notes (17 percent versus 11 percent) and unpermitted collaboration (49 percent versus 27 percent), among other forms of cheating.

A serious commitment to the honor code is crucial to its efficacy. As Professors McCabe and Treviño insist, an honor code should be “well implemented and strongly embedded in the student culture.”

What does that look like in practice? A few schools start the academic year with an actual commitment ceremony, where each student has to publicly pledge to uphold the school’s code. To this can be added a requirement to affirm the honor code on each graded assignment.

When I was an undergraduate at Princeton, every paper we turned in had to have the honor code written out and then signed. Now as a professor at Wake Forest, I make my class recite aloud with me before each exam our entire honor code and then sign it.

Signing an honor code can, among other things, serve as a moral reminder. As we know from both ordinary life and recent experimental findings, most of us are willing to cheat to some extent if we think it would be rewarding and we can get away with it. At the same time, we also want to think of ourselves as honest people and genuinely believe that cheating is wrong. But our more honorable intentions can be pushed to one side in our minds when tempting opportunities arise to come out ahead, even if by cheating. What a moral reminder does, then, is help to place our values front and center in our minds.

This is borne out by recent findings in the lab. In a widely cited study, Nina Mazar at the Questrom School of Business at Boston University and her colleagues had one group of students take a 20-problem test where they would be paid 50 cents per correct answer. It was a hard test — students averaged only 3.4 correct answers. A second group of students took the same test, but they graded their own work and reported their “scores” with no questions asked. The average in this group was 6.1 correct answers, suggesting some cheating. The third and most interesting group, though, began by signing an honor code and then took the test, followed by grading their own work. The result? An honorable 3.1 correct answers. Cheating was eliminated at the group level. Signing the honor code did the job.

Studies of honor codes and cheating have typically been conducted in face-to-face environments. But as we settle into the routine of online instruction, we should consider trying to extend the impact of an honor code virtually as well.

Honor codes won’t eliminate cheating. Deeply dishonest students will not be deterred. But fortunately, the research confirms what experience suggests: Most students are not deeply dishonest.

Christian B. Miller (@CharacterGap) is a professor of philosophy at Wake Forest University, the director of the Honesty Project and the author, most recently, of “The Character Gap: How Good Are We?”

The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.

Follow The New York Times Opinion section on Facebook, Twitter (@NYTopinion) and Instagram.