When I signed the honor code pledge at Princeton University for the first time, I must admit that I was skeptical. To someone who had observed several instances of blatant cheating in high school, Princeton seemed more than a little naive. Simply asking students to sign a pledge that went something like "On my honor as a gentleman, I give my word that I have neither given nor received aid during this examination," did not strike me as an effective deterrent. After all: I was sure at least some of my classmates had cheated in high school; the courses at Princeton were difficult; and our tests and examinations were unproctored, a new experience for most of us. Yet I never saw anyone cheat on a test during my four years at Princeton and I remember hearing talk of only two or three possible violations. Although I'm sure that more cheating than this went on, I remain convinced today that the honor code experience was one of the most valuable lessons my classmates and I received at Princeton.

Thus, when I decided to leave the corporate world after twenty years and become an academic, I was more than a little disappointed to hear the war stories of my MBA students when the topic turned to questions of student values and integrity. According to many of them, cheating was now a common occurrence. Because of my own experience as an undergraduate, I wondered whether cheating was as widespread as it seemed to them and whether honor codes still worked as I remembered. To find out I surveyed over six thousand students in the 1990-1991 academic year at thirty-one highly selective, small to medium sized, private colleges and universities across the United States.

Fourteen of the schools in this sample had academic honor codes. Although various definitions of honor codes have been proposed, I prefer the definition proposed by Brian Melendez in a study done at Harvard in 1985. Melendez suggested the defining characteristics of an honor code are one or more of the following: unproctored examinations, some kind of pledge requirement, a peer judiciary, and reportage-an obligation placed on students not to tolerate violations of the honor code by other students. All of the schools classified as code schools in my research met at least two of these criteria, and most met three or four.

The fundamental conclusion of my research is clear: there is significantly less cheating at schools with honor codes. Not only do fewer students at honor code institutions engage in any form of academic cheating (57% versus 78% of students at non-code schools), the power of honor codes is particularly clear when we look at students who repeatedly cheat on tests (or examinations). While 20% of the students at the non-code schools in my sample reported engaging in more than three instances of explicit test cheating, only 5% of the students at honor code schools did so.

A Most Important Question
More surprising, and perhaps more important, was the finding that the honor code itself did not appear to be the primary factor in explaining these differences in cheating. Rather, my data suggest that the most significant contextual influence on an individual student's decision whether or not to cheat is his or her perception of what other students on campus are doing. Not only does the observed behavior of peers seem to influence academic dishonesty, it also seems to provide a kind of normative support for it.

Indeed, an institution's ability to develop a shared understanding and acceptance of its policies on academic integrity appears to be a major influence on the level of cheating whether or not a school has an honor code. One of the lowest rates of self-reported academic dishonesty in my research was at a non-honor code institution. Though without an honor code, the institution is strongly committed to the concept of academic honesty, makes it a major topic of discussion for incoming students, and goes to great lengths to ensure its policy is understood and accepted as an obligation of every member of the campus community. The most important question to ask concerning academic dishonesty may be how an institution can create an environment where academic dishonesty is socially unacceptable, where institutional expectations are clearly understood and where students perceive that their peers are adhering to these expectations.

Comments from students at schools that appear to have achieved such a shared sense of responsibility for academic integrity seem to support this view. "I like the respect I get at [the institution] and wouldn't do anything to jeopardize that." "Peer pressure-you would feel very embarrassed if other students saw it." "... as for cheating on a test, it's socially unacceptable."

Comments from students at schools with high levels of self-reported cheating more often focus on rationalizations for cheating. "[Academic dishonesty] is rampant at ...., so much so that the attitude seems to be everybody does it-I’ll be at a disadvantage if I don't." "If others do it, you're being left behind by not participating." "When most of the class is cheating on a difficult exam and they will ruin the curve, it influences you to cheat so your grade won't be affected." (For a more detailed report of these results, see D.L. McCabe & A.K. Trevino, "Academic Dishonesty: Honor Codes and Other Contextual Influences," Journal of Higher Education, 64 (1993), 520-538.)

Declining Morals?
Although it is tempting to suggest that such negative attitudes simply reflect the declining social mores that we hear so much about, the data available provide mixed evidence. A comparison of my data with those collected by William Bowers in the 1960's suggests that any increase in student cheating has been fairly modest at the small-to-medium-sized, selective institutions in my original sample. The only dramatic change is a marked increase in the number of students who report working together on graded assignments even when the instructor has asked for individual work. At least on this issue, today's students feel free to define their own rules, believing that as long as learning takes place, it does not matter how. At larger, less selective schools, however, we see a different picture. The preliminary results of a study of 1,800 students at nine medium-to-large state schools, which I am now completing, suggests there has been a more significant increase in cheating at such schools.

As someone who teaches in a graduate school of management, one of my most disappointing conclusions was the fording that students who plan to enter business self-report more cheating than any other intended occupation. I think two factors are at play here. First, students headed for the corporate world often seem to apply a bottom-line mentality to their academic work; it doesn't matter how you achieve the result as long as you get the job done. Second, there seems to be a relation between cheating and the nature of testing. Marry students explained that it's easier to cheat on objective tests (those with only one right answer) than on the essay tests typically used in liberal arts courses. This may help us understand the higher rates of cheating among those preparing for careers not only in business, but also in engineering and medicine, where science courses dominate the undergraduate curriculum.

An interesting question is whether honor codes, or other institutional strategies to address academic dishonesty, influence an individual's ethical behavior later in life. Although there is anecdotal evidence that suggests they may, there is not much empirical research on this question. I am now looking at this question. Some colleagues and I recently surveyed a group of alumni from two highly selective liberal arts colleges in the northeast who work in business. One of these schools has had an honor code since the late 1800s; the other has never had one. Although we did not ford a significant difference in the ethical behavior of these two groups, our initial results do suggest that those with honor-code backgrounds were significantly influenced by a strongly implemented corporate code of ethics in their work place while the presence of a corporate code did not have a significant influence on those without a collegiate honor-code experience. This result suggests that there may be interaction between a collegiate honor-code experience and a strongly implemented corporate code of ethics. Why? One possible explanation is that an honor-code experience makes individuals more open to the guidance provided by a strong corporate ethics code.

Honor and Education
Although more research is clearly needed to confirm this relation, I remain convinced that the basic elements of an honor code can be an important part of the undergraduate educational experience. Not only do they help reduce cheating, they may also teach students a more valuable lesson.