Cheating Lessons

James M. Lang

• Publisher: Harvard University Press
• Year: 2013
• ISBN:
  9780674727304
• Topic: Education
• Subtopic: Higher Education

Nearly three-quarters of college students cheat during their undergraduate careers, a startling number attributed variously to the laziness of today's students, their lack of a moral compass, or the demands of a hypercompetitive society. For James Lang, cultural or sociological explanations like these are red herrings. His provocative new research indicates that students often cheat because their learning environments give them ample incentives to try—and that strategies which make cheating less worthwhile also improve student learning. Cheating Lessons is a practical guide to tackling academic dishonesty at its roots.Drawing on an array of findings from cognitive theory, Lang analyzes the specific, often hidden features of course design and daily classroom practice that create opportunities for cheating. Courses that set the stakes of performance very high, that rely on single assessment mechanisms like multiple-choice tests, that have arbitrary grading criteria: these are the kinds of conditions that breed cheating. Lang seeks to empower teachers to create more effective learning environments that foster intrinsic motivation, promote mastery, and instill the sense of self-efficacy that students need for deep learning.Although cheating is a persistent problem, the prognosis is not dire. The good news is that strategies which reduce cheating also improve student performance overall. Instructors who learn to curb academic dishonesty will have done more than solve a course management problem—they will have become better educators all around.

PART ONE

BUILDING A THEORY OF CHEATING

Tricia Bertram Gallant, one of the lead researchers on cheating in higher education today, offers an excellent and concise overview of academic dishonesty from 1760 to the present day in her book Academic Integrity in the Twenty-First Century. In that history you will find an oft-repeated gem of a quotation from an Ivy League administrator in 1928:

Many young men and women who are scrupulously honorable in other relationships of life seem to have little hesitancy in submitting themes and theses which they have not written, in bringing prepared ‘cribs’ to examinations, and in conveying information to one another during the course of an examination. There is a not uncommon feeling that a state of war exists between faculty member and students . . . partly explained by the unsympathetic attitude of some professors, and partly by the rather mechanical organization involving grades, warnings, and probation; but, certainly, the principal cause must be found in the failure of undergraduates to appreciate the value to themselves of serious and conscientious intellectual effort and achievement.¹

If you come to the problem of cheating in higher education expecting to find evidence that we are in the midst of a cheating epidemic, and that the problem is much worse now than it was in the idyllic past, this quotation will give you pause. The statistics I will present in Chapter 1 should give you even greater pause, and I hope will help convince you that cheating and higher education in America have enjoyed a long and robust history together.

But if you look carefully at this quotation, you will also see in the words of this beleaguered administrator an acknowledgment of the extent to which the learning environments on his campus help exacerbate the problem. He points both to the “mechanical organization” of courses and to the “unsympathetic” attitudes of the professors who teach them. One can easily envision the types of courses to which he refers: the sole responsibilities of the students are listening to lectures, taking exams, and bemoaning their grades; faculty members, by contrast, are both dispensers of knowledge and judges who have no sympathy for students who cannot meet the standardized expectations. While the writer apportions blame to the students for their “failure . . . to appreciate the value” of their educational experiences, one could just as easily point the finger back to the institution and argue that it has failed to convince the students of the value of what it has to offer. In either case, a chasm has opened between the faculty and the students in this description, and that chasm stems at least in part from the extent to which students are uninspired to learn, feel challenged instead of helped by their professors, and see their courses as stumbling blocks instead of steps to a better life.

This might strike some readers as an apt description of today’s educational environment, in which the war between students and faculty seems to have evolved into new forms. Yesterday’s cheat sheets have become today’s smart phones, and the majority of our students still don’t come to college with a full appreciation of the “value to themselves of serious and conscientious intellectual effort and achievement.” So if you find yourself occasionally frustrated at the reluctance of your students to embrace your courses or your discipline for the pure love of learning, as perhaps you did as an undergraduate, you can rest assured that our faculty forebears felt the same way. Fortunately for us, we have the benefit of several decades of research on how and why students cheat, and that research—much like the quotation above—tells us that the learning environments we construct can have an important influence on academic dishonesty on our campuses. In the three chapters of Part I, I delve into the research on cheating in higher education today, as well as cheating in more generalized learning or performance environments, in order to construct my own theory about the five features of our classrooms and campuses that have the greatest influence on how and why students cheat—and that may also have the greatest influence on whether or not our students learn.

1

WHO CHEATS—AND HOW MUCH?

In the spring of 1962, a doctoral student at Columbia University set out to create the first large-scale estimate of cheating rates in America’s colleges and universities. William J. Bowers did not have much to work with in the way of precedents for this research; some previous scholars had attempted to understand the psychological makeup of student cheaters, and a few others had attempted institution-specific studies to gauge cheating rates or investigate possible methods for reducing or preventing cheating. The goal of his project, Bowers explains in his introduction, was “to combine into a single research effort three objectives of previous research—identifying sources of cheating, evaluating remedies, and estimating rates.”¹ He began his project by sending questionnaires to the deans of students and student body presidents at “all regionally accredited colleges and universities across the country.” He used the responses he received from these questionnaires—which were returned by more than a thousand deans and student body presidents—to construct a survey instrument that was then mailed out, in the spring of 1963, to students at ninety-nine colleges and universities in the United States. The sample of schools in the survey is extremely diverse, including everything from Ivy League schools (Yale) and large public universities (Eastern Michigan) to religious institutions (Notre Dame) and small private schools (Reed).

In the end he received more than five thousand responses to the survey, which makes it far and away the first comprehensive effort to assess cheating rates in higher education in the United States. In the manuscript which he eventually published about his survey, Bowers uses the detailed demographic data that he collected from the students in order to estimate cheating rates for the general student population, as well as to analyze a wide range of what some cheating researchers call dispositional factors—in other words, individual features of your personality or your life situation that might dispose you to cheat. So in his diligent attempt to ferret out potential dispositional factors, Bowers provides statistical comparisons for just about any aspect of a student’s life you can imagine, including the cheating rates of those who dated frequently versus those who didn’t, those who spent a lot of time drinking and playing cards versus those who engaged in more wholesome activities, and those who were in college primarily to secure a spouse instead of to learn or improve their job prospects.

Bowers used three different means to try to assess the global rates of cheating among all students in his survey, each of which yielded different results. This fact has made the nutshell conclusion of Bowers’s work a difficult one to pin down, and explains why contemporary researchers give varying numbers when they refer to the results of his survey. But most frequently you will see academic integrity scholars citing the second of his three questions, which asked students to indicate whether they had engaged in thirteen very specifically defined behaviors that most of us would consider academically dishonest. Some of these behaviors are the worst sorts of cheating activities that we can imagine, such as “Having another student take an exam for you,” “Writing a paper for another student,” or “Copying answers from a text or other source instead of doing the work independently.”² Other described behaviors are ones that we would all likely agree are academically dishonest, but less serious than some of the others, such as “Getting questions or answers from someone who has already taken the exam” or “‘Padding’ a few items on a bibliography.”³ When Bowers gave students these specific descriptions of academically dishonest behavior, 75 percent of them admitted to engaging in at least one of them over the course of their college career.

If all you know about cheating in higher education today comes from the occasional scandal making its way onto the front pages, or from op-ed pieces or blogs on the subject, this number should come as a surprise to you. The most common plotline you will encounter in contemporary stories or essays about cheating is one which suggests that cheating rates are sky-high and rising. Glancing back at the statistical work of William Bowers can help us see very clearly that this picture gets it exactly half wrong: cheating rates may be sky high—depending on your definition of that unscientific term, of course—but they are not rising. Fifty long years ago, back when our forebears were trading in their bobby socks for bell bottoms and worrying over our entrance into a complicated war in a faraway land, 75 percent of students at America’s colleges and universities had cheated at least once in their college career.

Not much has changed since then, at least in terms of the global cheating numbers that concern us here. This will be easy enough for me to demonstrate because the lineage of prominent cheating researchers in higher education passes directly from William Bowers to Donald McCabe, who has been producing large-scale surveys in this area since the early 1990s, beginning just a few years after he left his job as vice president of marketing and sales at Johnson and Johnson for a faculty position at Rutgers Business School. McCabe’s early publications suggest a researcher still trying to find his way; titles like “The Measurement of Environmental Volatility” and “Making Sense of the Environment: The Role of Perceived Effectiveness,” both published in conference proceedings, give no hint of the long and productive research agenda that McCabe would eventually craft. The first entry in that agenda came in 1991, when he published an essay entitled “Context, Values, and Moral Dilemmas: Comparing the Choices of Business and Law School Students” in the Journal of Business Ethics in 1991. Something about analyzing the moral dilemmas of college and university students must have lit an intellectual fire in McCabe, who would proceed, over the course of the next two dozen years, to publish almost fifty articles about student cheating, academic integrity, and honor codes in higher education. In the early 1990s he helped co-found the independent organization the Center for Academic Integrity (now the International Center for Academic Integrity, affiliated with Clemson University) and served as its first president.⁴ While you can find plenty of published surveys of cheating behaviors in the literature of this particular subdiscipline, you will not find anyone whose name appears so frequently as McCabe’s, or whose work stretches into so many corners of the field.

McCabe’s incredibly prolific publication record draws from multiple large-scale surveys he has conducted, frequently in collaboration with other researchers, on cheating among college students. Until 2012, McCabe’s body of research and data interpretations had appeared in a very wide variety of formats and publications, all of them short form, from statistical reports in specialized academic journals to more general summary accounts for popular academic magazines like Change or the Chronicle of Higher Education. In 2012, though, McCabe and two of his frequent collaborators published a monograph, Cheating in College: Why Students Do It and What Educators Can Do About It, which provides a handy overview of his entire body of work, stretching from his first surveys in the early 1990s—conducted through the good old U.S. mail—to his most current series of web surveys. I will begin my analysis of contemporary cheating rates with a survey that McCabe and one co-author, Linda Klebe Trevino, conducted in the early 1990s which was designed to replicate Bowers’s 1963 surveys. From there, I will move forward to the overview statistics that McCabe provides us in his recent book about his most current survey results; these two sets of studies give us a thirty- and (almost) fifty-year perspective on Bowers’s findings and should help us answer the two pressing questions that most faculty have about cheating rates: How much are students cheating today? And are they cheating now substantially more than they did in the past?

In the fall of 1993, McCabe and Trevino mailed surveys to 500 students at each of nine different institutions, all of which had been included in the Bowers study. They cast a much narrower net than Bowers did, however, in selecting the type of institution from which to draw their survey data: all nine of them were “public institutions of higher education ranging in size from just under 5,000 students to over 35,000 students, with a mean undergraduate enrollment of 12,329 students.” They describe the institutions in their sample as “moderately selective in their admission criteria.” The total number of students responding to the surveys ended up at around 1,800.⁵ In response to our second question, about whether global rates of cheating had increased since 1963, McCabe and Trevino provide a succinct statement of their findings in a 1996 essay in Change magazine: “The dramatic upsurge in cheating heralded by the media was not found.”⁶ In a later summary of their findings published in the academic journal Ethics and Behavior, McCabe and Trevino (along with a third researcher, Kenneth D. Butterfield) establish a category of “serious cheating” to cover the thirteen or so cheating behaviors addressed in both studies, and cite a summary rate of 75 percent in Bowers’s 1963 survey versus an overall rate of 82 percent in the 1993 replication.⁷

You may be thinking to yourself that a 7 percent increase, while not huge, certainly counts for something. So let’s move forward from that 1993 survey, with its small bump in the overall cheating rate, to the more recent conclusions that McCabe and his colleagues have drawn about cheating rates in higher education, as documented in Cheating in College. The most recent data set that McCabe covers in this book comes from an extensive series of web surveys conducted from 2002–2010. Although these surveys come from an eight-year period and cover a huge number of students—almost 150,000 from the United States and Canada—from a range of institutions, McCabe and his colleagues consider them as a single dataset. And that dataset speaks quite clearly about the direction in which cheating rates have moved in the period since their 1993 replication. As McCabe and his colleagues explain, “self-reported cheating recorded in the 2002–2010 web surveys is lower than in any previous surveys.”⁸ In the table that McCabe provides to allow readers to see comparative cheating rates in nine different categories, and which concludes with an overall rate of cheating in his most recent surveys, the rates of cheating come only from schools that do not have an honor code (more on this later). When the honor code schools are removed, the Bowers 1963 survey yields an 83 percent rate of overall cheating; the 1993 replication by McCabe and colleagues yields a rate of 87 percent; the cheating rates for 2002–2010 clock in at 65 percent. According to McCabe’s research, then, cheating rates rose somewhat into the 1990s, and then have actually been dropping quite substantially over the course of the past decade.

McCabe and his co-authors, however, are wary of accepting these numbers at face value: “Though we would like to believe that this is an accurate assessment of prevalence,” they explain, “we have several reasons to be skeptical.”⁹ Those reasons stem, in part, from the fact that all cheating surveys, from Bowers to the present day, are based on student self-reports of their cheating behavior. As the authors of another book, Cheating in School, point out, “we really know what students claim to be doing, rather than what they are actually doing.”¹⁰ And this introduces uncertainty in the numbers in a variety of ways. First of all, it depends upon students trusting that their answers will not be revealed to their teachers—or, as McCabe and his colleagues label it, “concerns about the confidentiality of the electronic data.”¹¹ I suspect that students who put anonymous surveys in the mail to some faraway researcher were more willing to trust the confidentiality of their responses than students who are filling out surveys online, and who are well aware of the many ways we have, in the twenty-first century, to track the identities of anyone with an internet connection. So student self-reporting of cheating behaviors may be lower on the current web surveys for that reason.

Second, and perhaps more substantively, cheating self-reports rely upon the students having a clear understanding of what constitutes cheating—and an understanding that correlates with the survey administrator’s understanding. Susan Blum, like many other researchers in the area of plagiarism, has pointed out the extent to which today’s students have difficulty in distinguishing between acceptable and unacceptable citations from the work of others, in part because of the rise of intertextuality in their lives: “student citation norms differ considerably from academic citation norms. Students accept that in everyday conversation, quotation is fun and playful, though certainly not obligatory; they provide citation only when necessary to direct their peers to rare or unfamiliar sources; they regard the cutting and pasting of pastiche as ...

Table of contents

• Cover
• Title Page
• Copyright
• Dedication
• Contents
• Introduction
• Part One. Building a Theory of Cheating
• Part Two. The (Nearly) Cheating-Free Classroom
• Part Three. Speaking about Cheating
• Conclusion: The Future of Cheating
• Notes
• Acknowledgments
• Index