Skip to content

Why We Cheat

Good people make bad decisions when they convince themselves they deserve a break from the rules.

Illustration: Scott Bakal

By Joan O'C. Hamilton

It was once a common schoolyard taunt, delivered with a snarling, singsong cadence: "Cheaters never PROSper! Cheaters never PROSper!" The perp, having been caught in the act of copying from a neighbor's paper, would reflexively hiss: "I am not a cheater!"

That shaming ritual—and the response—offers insights into some dark but fascinating puzzles. Fifty years of research by social scientists at Stanford and elsewhere shows that the majority of people who cheat don't see themselves as cheaters. Whether the context is a classroom, a locker room or a carpool lane, otherwise law-abiding people who circumvent the rules rationalize their choice to cross a line by redrawing the line. Says Benoit Monin, professor of organizational behavior and psychology at Stanford, "It's clear that people cherish the idea of being a moral person, so the question is: When does cheating become acceptable? There seems to be a paradox that people do things that are not great, but then they still look at themselves in the mirror and think they see a good person."

Cheating is a complex and ancient topic. Adulterous temptations have liquefied moral backbones since biblical times. These days, the offenses that betray a promise, give an unfair advantage or undermine the integrity of an institution occur in weird and extravagant ways. Even the 2015 Super Bowl-winning New England Patriots quarterback Tom Brady now wears the scarlet C thanks to "Deflategate."

A rash of cheating scandals at leading universities has again aimed a spotlight on the issue and renewed interest in understanding what motivates people to flout the rules. Researchers have found that even excellent students can be tempted to cheat when certain conditions line up. When we're feeling tired or overwhelmed; when the likelihood of getting away with it is high; and when there is a perception that others are cheating, too, it can be easy to rationalize taking the low road.

Monin says two conditions clearly can motivate cheating among high-achieving students: first, the perception that "everybody else is cheating, so it can't be that bad;" second, the worry that "I'll fall behind unless I cheat." (The same reasoning was used to explain steroid use by baseball players and blood doping by professional cyclists.) The psychological underpinnings of cheating are important, but administrators, researchers and ethicists are trying to understand whether other factors—technology, parenting styles and pressure-cooker environments on campuses—are driving more cheating today.

In May, following news that 130 students were caught copying code in computer science courses, Provost John Etchemendy, PhD '82, sent a letter to all Stanford faculty reminding them of their duty to make the importance of academic integrity clear. Similar reports of student cheating have bedeviled Harvard, where in 2012 dozens of students were accused of, and many of them ultimately suspended for, inappropriately collaborating on a take-home test. Other variations on the theme have emerged at Dartmouth, the University of North Carolina and the Air Force Academy. "The ease with which one can run afoul of community standards seems to be increasing," Harvard's Dean of Undergraduate Education Jay D. Harris told a campus forum in 2013. "This instinct is as old as humanity itself," he figures, but "the instruments to bring these things about are more available and are making it easier."

Surveys suggest that many high school and college students see cheating as a minor offense. However, Monin wonders if the evolution of higher education itself, with many more technologies in the mix, may make both cheating and identifying cheating easier. For example, Stanford computer science students are warned on day one of their classes that their assignments will be subjected to a computer program that looks for similar coding between students. That is how the Stanford students ended up being reported last spring. Did they just not care about the consequences? It's not that simple, Monin observes. The specific context and confusion over how much "sharing" of code is either acceptable or identifiable can play a role in a given decision to cheat. "It's easy to say, 'God, 20 percent of the students in this class are cheaters,' but I would not assume these people are cheating in other classes."

Susan Fleischmann, associate dean in the Office of Community Standards, which administers Stanford's Honor Code, is not convinced there is a big spike in cheating. In 2008-09, OCS processed 134 cheating cases. Between 2008-09 and 2013-14, the number of cases never exceeded 130. This year, the case number ballooned to 260, but as Fleishmann points out, half of those came from the computer science incident.

In most cases, it's a faculty member who brings cheating allegations to OCS. A student accused of an Honor Code violation can opt either for "early resolution" and take immediate responsibility for the infraction and the consequences (probation or suspension), or the student can challenge the accusation. Fleischmann says about 75 percent of students choose early resolution, and the rest proceed to a hearing before a judicial panel of faculty, staff, and students. Fleischmann adds that Honor Code cases involve less than 1 percent of students every year. "A lot of students are getting by just fine without violating the Honor Code." (See sidebar.)

Fleischmann is well acquainted with the "I am not a cheater" defense. However, she does not believe students in the cases she sees are serial cheaters. "Students do rationalize because they want to maintain an image that they have of themselves, or that others have of them. But I believe the majority of students do not intend to cheat. I think the pressures get to them and they make a bad decision."

Alexandra Bourdillon, a sophomore from Cupertino, was not involved in the computer science class situation, but she understands the pressure to cut a corner. "My impression is that people cheat when they are overwhelmed by the work. In computer science you can have projects where you spend five or six hours working on it night after night. If you hit a bug, you tack on another hour per bug to fix that, and now you haven't touched your other classwork for three days." She agrees with other students who say the lines differentiating valid collaboration from improper copying are blurry, and vary from one class to another. "Some students who haven't had computer science before might think it's OK to share just a small part of the code, but it isn't."

Technology also enables new ways to cheat beyond just tech courses, from buying essays online to easy sharing of data with other students via text or email to unauthorized use of the Internet during tests. At Dartmouth, 43 students in the fall 2014 Sports, Ethics and Religion course were sanctioned for exploiting a digital "clicker" system used to allow students to answer questions from their seats during class (theoretically to motivate better attendance). Eventually, the instructors realized they were getting more responses than there were bodies in the classroom. It turned out students cutting the class would just give their clicker to another student who answered for them.

Obviously, pressure to succeed is not confined to undergraduates. In a widely reported paper published in Science last December, a UCLA graduate student claimed his study showed that voters who spoke with a gay canvasser about gay marriage changed their views to support gay marriage. But the study's co-author, Columbia University professor Donald P. Green, later asked Science to retract the paper when the student admitted he had misrepresented his research methods and was unable to produce the data on which the paper was based. It added to the list of research fraud cases that have been unmasked in recent years, some of which had dangerous or even lethal consequences connected to pharmaceuticals or risky medical procedures.

Cheating - inline 1
Illustration: Scott Bakal

The study of the thought process involved when people cross ethical and moral boundaries goes back decades. One line of research examines so-called cognitive dissonance, coined at Stanford in 1957 by psychology professor Leon Festinger. Cognitive dissonance is the disquiet a person feels when balancing or trying to process two contradictory situations. The result can be elaborate rationalizations. For example, "I am a good person" and "I cheated" aren't easily reconciled. Doing so requires changing one's self-image, which most people resist, or altering one's view of one's behavior—i.e. "it wasn't really cheating."

In 1958, in another frequently cited study, Stanford psychology graduate student Judson Mills, PhD '58, tempted a group of sixth graders to cheat. Before the experiment, he polled the children and learned that some were staunchly anticheating, while others didn't see cheating as all that serious. He then assigned them tasks in which he made it easy to cheat, and he made it appear that their cheating could not be detected. The next day he polled the children again. Students who had said cheating was bad but cheated anyway now said cheating wasn't really so bad. They rationalized that because they had committed the very act they had said was terrible, it must not be so terrible after all.

Duke University researcher Dan Ariely is a professor of behavioral economics who wrote the 2012 book The (Honest) Truth About Dishonesty. Ariely notes that for many years the dominant school of thought about cheating followed the theories of Nobel Prize-winning economist Gary Becker. Becker argued that people rationally analyze their decisions, weigh the pros and cons, assess the risks and penalties, and land on a choice that produces the best outcome for the lowest cost.

Ariely decided to test that. He created experiments involving math problems in which subjects were tempted by greater or lesser rewards, and higher and lower chances of getting caught, and in which the behavior of others around them informed their mindset. Based on his results in one experiment, 70 percent were tempted to cheat at least a little. But it turns out the value of the reward plays a lesser role than Becker would have predicted. Instead, people seemed to calculate the largest benefit they could get for a transgression that would still allow them to tell themselves they weren't a cheater. "Essentially, we cheat up to the level that allows us to retain our self-image as reasonably honest individuals," Ariely says.

Self-image played a big role in a 2013 study by Stanford's Monin and colleagues Christopher Bryan, PhD '09, and Gabrielle Adams, PhD '11, results of which were published in the Journal of Experimental Psychology. Participants were given a chance to cheat in self-reporting their performance on a task. One group was admonished "Please don't be a cheater," while the other was told "Please don't cheat." The first group consistently cheated less than the second. The researchers concluded that invoking people's desire to maintain a good self-image—in this case simply by avoiding the label of a cheater—can motivate better behavior.

Monin wonders if the motivation to cheat is intensifying because the competition to secure admission to and then excel at top universities has crossed some kind of pressure threshold. "Not to excuse any of it, but it may be partly our fault. We place the bar higher and higher; we want the perfect transcript."

Fleischmann agrees that students seem to "get desperate when they're staying up until all hours and they can't fathom getting less than an A. The pressure they're putting on themselves is too much. They have been used to having so much support in high school, but (in college) they don't have it."

Or they may have support that inflames the stress. So-called helicopter parenting and a focus on self-esteem above all can make students feel they must live up to an image. Research by Stanford psychology professor Carol Dweck suggests that a student who has been raised to believe that his or her success stems from special abilities rather than hard work and learning from mistakes may become fixated on maintaining that "gifted" status at any cost, which could include cheating.

Fleischmann says aggressive interventions by parents defending their children in cheating cases appear to be on the rise everywhere. "Even five years ago we didn't have attorneys involved and we didn't receive as many calls from parents," she says.

So, do cheaters never prosper? Clearly some, possibly even most, do. But it also appears we are wired to balance our transgressions with our selfless acts and good behavior. One of the intriguing mysteries, Monin explains, is in the realm of "moral licensing." That is, when does past "good" behavior liberate one to push further into "bad" behavior? It's not hard to imagine a student rationalizing that if she spent the weekend tutoring low-income children instead of studying, cheating on the test doesn't make her a cheater.

The problem is that regardless of motivation, acts of cheating chip away at the core mission of a university. In his note last spring, Etchemendy reminded Stanford faculty that "dishonesty is corrosive in an academic community." Every bogus study or fudged data set and every grade point average that lacks integrity eat into the value of the education students receive, but they also diminish the credibility of the degree they've "earned." And that's something that affects every student, whether or not he or she is cheating.

Joan O'C. Hamilton, '83, is a regular contributor to the magazine.

Comments (1)

  • Mr. Thomas Hwang

    I find it odd that students are encouraged to collaborate and help each other regarding doing homework, writing papers, sharing class notes, sharing insights.  But then when a take home test is issued, the rules change.

    For an in class test, the rules are pretty clear.  

    But if you don't want cheating on take home exams and certain writing essays, then the University needs to stop sending mixed messages.

    For me, the answer is easy.  Get rid of take home tests.  

    Posted by Mr. Thomas Hwang on Sep 9, 2015 6:57 PM


Your Rating
Average Rating



Be the first one to tag this!