Academic journal article College Student Journal

Explaining Variations in the Level of Academic Dishonesty in Studies of College Students: Some New Evidence

Academic journal article College Student Journal

Explaining Variations in the Level of Academic Dishonesty in Studies of College Students: Some New Evidence

Article excerpt

Numerous claims have been made in the popular press that the level of student dishonesty in colleges and universities is increasing. This paper reports a study that investigated the relationship between the overall level of cheating reported in studies of college students and the year of publication of the study, as well as three other independent variables: sample size, the type of study (survey v observational), and the number of practices included in the study. Our findings do not support claims that the level of student dishonesty has increased. Only the number of practices included in the study was found to be related to the overall level of student cheating. We conclude that comparisons of studies that included different numbers of unethical academic practices must be made with caution.


Several explanations have been offered for the differences found in the level of academic dishonesty in studies of college students. One of these is that the level of academic dishonesty has increased over time. This explanation has received considerable attention in the popular press (for examples, see: Donahue & Heard, 1997; Kleiner & Lord, 1999; "Your cheatin' heart," 1992), but has not always been supported by research.

Baird (1980) cited five studies conducted between 1941 and 1970 that showed a change in the cheating rate from 23% to 55%. In his 1980 study he found that about 75% of undergraduate business, liberal arts, and education majors had cheated in college. He concluded that the cheating rate in college had been increasing, and that his data showed a continuation of the upward trend.

McCabe (McCabe & Bowers, 1994) conducted a broad-based survey of college students in 1990. He extracted a sub-sample from his database that matched the sample used in a 1962 study by Bowers and compared the rates of participation in nine unethical academic behaviors included in both studies. The subjects were junior and senior males in small to medium sized residential schools that had selective admissions policies. In non-honor-code schools, collaboration on individual work increased, while cheating on tests, plagiarism, and turning in work done by others decreased. In schools with honor codes, cheating on tests and collaboration increased, while other forms of cheating decreased. The authors stated that the high rates of increase in cheating heralded by the media were not found.

In 1993 McCabe and Bowers (McCabe & Trevino, 1996) surveyed students enrolled at nine medium to large state universities that were in Bower's 1962 sample. In the Bowers study, 63% of respondents admitted cheating in college. The rate was 70% in 1993. Rates of cheating on exams and collaboration on individual work increased. The rate of copying from another student's exam went from 26% to 52%, while collaboration increased from 16% to 27%. However, plagiarism and turning in work done by someone else decreased slightly.

Cole and McCabe reported in 1996 that surveys of undergraduate students at Stanford in 1976, 1980, and 1984 found no significant changes in types or levels of student dishonesty. In a second study published in 1996, Diekhoff (Diekhoff, LaBeff, Clark, Williams, & Haines) found some increases in student dishonesty from 1984 to 1994 in two first year survey courses. Cheating on exams stayed about the same, but cheating on quizzes increased from 22% to 31%, cheating on assignments increased from 34% to 45%, and the proportion of students cheating overall increased from 54% to 61%.

Spiller and Crown (1995) identified 24 studies conducted between 1927 and 1986 that operationalized cheating as college students' changing answers on self-graded tests. They excluded from their sample studies where "artificial means" were used to raise the percentage of students who might cheat and where cheating was measured over multiple examinations. They used linear regression to examine the influence of time on this one form of college cheating, where the percentage cheating was the dependent variable and the year the study was conducted was the independent variable. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.