Last week, Harvard University’s campus newspaper published a summary of a survey administered to the incoming first-year class, the results of which included shocking statistics about the freshman at America’s most prestigious institution of higher education. “Ten percent of respondents,” the Harvard Crimson reported, “admitted to having cheated on an exam, and 17 percent said they had cheated on a paper or a take-home assignment. An even greater percentage—42 percent—admitted to cheating on a homework assignment or problem set.”
This comes a year after half of the 250-person class in government at Harvard was accused of cheating on the take-home final. That scandal was almost immediately followed by one in which the administration, while scrounging for evidence on who had leaked the story to the press, secretly searched the e-mail accounts of more than a dozen resident deans.
(MORE: The Problem with Prize Culture)
Last week’s survey results will provoke the predictable bouts of hand-wringing among worried academics, and speculation in the media about the rise of cheating among today’s college students. If the pundits stay true to form, they will argue that those Harvard students are a) inspired by today’s technologies to new heights of academic dishonesty, or b) imitating the dishonest behaviors of their parents and role models, or c) reflecting the larger decline or morals and values in America today. Or all three.
None of these depressing conclusions, however, are consistent with the historical record of cheating in higher education. The first major survey on cheating rates at America’s colleges and universities was conducted in 1963 by William J. Bowers, a doctoral student at Columbia University, who asked more than 5,000 students at institutions around the country whether they had engaged in any of thirteen specifically-defined cheating behaviors. 75% of them admitted to cheating at least one time in their college careers.
(MORE: Over-Practicing Makes Perfect)
Subsequent cheating researchers have found little movement in these numbers over time. In fact, an extensive series of surveys conducted by Donald L. McCabe of Rutgers University and a team of researchers from 2002-2010, and reported in his book Cheating in College, found one-time cheating rates among America’s college students hovering between 60-70%—or a bit lower than the numbers Bowers found in 1963.
In other words, cheating levels are fairly high, but they have always been so. The better question to ask is why. Duke University researcher Dan Ariely and his colleagues have conducted dozens of experiments designed to see what makes people willing to engage in acts of cheating and dishonesty in their everyday lives. Their findings have been remarkably consistent: most people, under the right circumstances, are willing to engage in small acts of dishonesty. This seems to be a part of our human nature. With enough incentives in front of us, most of will cheat at least a little bit. Fortunately, Ariely’s formulation has a happy corollary: Since most people are willing to cheat under the right circumstances, our best approach to reducing cheating will be to change those circumstances.
(MORE: Is School Just For Getting A Good Job?)
If we open our minds to this possibility, we will have to reconsider the nature of the learning environments we are presenting to our students at both the high school and college level. When we do so, we might find that a significant portion of college and university courses feature precisely the types of circumstances that induce cheating. For example, when courses rely on infrequent, high-stakes assessment exams—two midterms and a final exam—they put intense pressure on a small number of opportunities for students to earn their grades, and hence ratchet up the incentive to cheat.
But, and more importantly, students learn less from courses with such infrequent, high-stakes assessments. In order to achieve deep and meaningful learning, our students need frequent practice in recalling and working with concepts and information, as Annie Murphy Paul has rightly argued. Two midterms and a final do not provide them with that required practice. Change those circumstances, and we will reduce cheating and induce learning.
In the wake of the new Harvard survey, we can choose to wring our hands, or condemn our students for their bankrupt values, or lament the presence of technologies in our classroom. But we would be much wiser to see academic dishonesty, and the research on what causes it, as offering us lessons in how to build better learning environments for our students.