A professor at the Rutgers Business School is out to discover just how much college students cheat and why.
Donald McCabe - a professor in management and global business - is conducting the third round of his national survey of college students, and Tufts is among the 28 schools participating.
McCabe previously conducted the survey in 1990 and 1995. Tufts was also part of the survey before.
"When this work was originally done ten years ago," Dean of Students Bruce Reitman said, "the results for Tufts were similar to that of the other institutions. But they were very surprising in that close to half the students admitted that they had cheated on three or more occasions."
Reitman encouraged this year's sophomores, juniors and seniors to respond to McCabe's voluntary survey in an e-mail Nov. 7.
This year's survey features questions on the University's penalties for cheating, the clarity of the policies, and the frequency of cheating.
McCabe first conducted the study 15 years ago, using a questionnaire developed by William Bowers for his doctoral dissertation at Columbia University. McCabe has made changes to the questions based on his results from previous surveys.
"Sometimes students or administrators give me ideas and I keep on revising it," McCabe said. "But I don't make revisions unless they make sense with the data I've collected."
McCabe has changed specific questions and varied the length of the survey, but he said the current survey is similar to the one from ten years ago.
"Basically I'm trying to find out students' attitudes about academic dishonesty and what kinds of things students are doing," McCabe said. "I'm trying to identify areas that would be the most fruitful for students to work on."
One of the things that has increased since Bowers' work in the 1960s is the likelihood of students to work together when they are asked for individual work, McCabe said. Students give many reasons for their decision to work together even when asked not to, he said, including that it is easier to learn together and that they do not have enough time to do the work on their own.
"Hopefully these results could guide different policies in the classroom to eliminate this trend," McCabe said.
His research has also shown a correlation between students' grade point average and their likelihood to cheat.
"It's kind of a U-shape relationship," he said. "Students at the bottom of the GPA spectrum, the ones that are really struggling, maybe they're on an athletic scholarship, tend to self-report more cheating. But then the high-end hypercompetitive students, maybe the ones striving for a doctorate, that will only accept Harvard and Stanford, also report a fair amount of cheating."
"You see a little bit more at the bottom but it's at both ends," McCabe said.
The questions in the last survey outlined the definition of cheating - on exams, class work, projects or any form of graded material. "It was a bigger deal than we had expected," Reitman said. "At that time there was a great deal of interest in honor codes and academic integrity and whether they were diminishing and the research addressed that. It's of great interest to see if over the past ten years things have changed."
Each of the 28 schools participating in the survey this year participated previously. Of the schools, 14 are traditional honor code schools, and 17 are not. An honor code school is a school where students sign a pledge that they will do their own work honestly, report students who are cheating, and observe a high standard of behavior. Exams are typically not proctored and are often take-home.
While McCabe's results from ten years ago showed that honor code schools demonstrated a little less academic dishonesty than non-honor code schools, the system does not always work well, Reitman said.
Reitman said it was unlikely Tufts would adopt an honor code system. "It's difficult to suddenly become an honor code school because you can't change people's attitudes now that they're here," he said. "An honor code school is something that students have to look for and apply to in advance with the knowledge they will be committing to that standard."
McCabe's survey is voluntary, which could lead to some inaccuracies in the results. Typically only 15 to 20 percent of the students at a particular school respond to the survey, he said. At all of the schools combined, about 5,000 to 6,000 students respond.
"Suppose I found out that 92 percent of the students at Tufts reported cheating on a test," McCabe said. "I wouldn't trust that data or say it is true of all the students at Tufts. There's a lot of reasons why people wouldn't want to self-report their cheating."
He said, though, that if a certain amount of students are willing to report their cheating, he assumes there are even more who are not reporting it.
McCabe also considers it important to compare students from different groups, such as athletes versus non-athletes, and high GPA versus low GPA.
"I think those comparisons are informative but in terms of taking the exact level of cheating I don't think the survey is highly reliable," he said. "But if for example I found that the average cheating was 30 percent, and at Tufts it was 60 percent, I'd say, 'Tufts, you have a problem.'"
McCabe said the biggest effect of his research has been on increasing awareness of cheating, though he said, "I don't know if I've been successful in helping to reduce cheating."
He started the Center for Academic Integrity at Rutgers to make research on cheating available to schools, host conferences and help schools develop academic integrity policies. The Center is now at Duke University and has over 390 member high schools and colleges.
"I think there's a lot of evidence that ethics and morals in general have declined in the past few decades," he said. "I'm trying to see if it's possible to reverse that trend on college campuses."
The survey at Tufts will last for three of four weeks, McCabe said. This year's effort began a month ago, and all the schools are expected to have completed the survey within about 45 to 60 days.
Tufts is one of only four schools doing the survey only online. The other 24 have a written component, which McCabe said makes the process take longer. Four are only doing the survey on paper, and 20 are doing the survey both on paper and online.
Reitman said he was looking forward to McCabe's report. "Is academic honesty an issue at Tufts?" he asked. "I'd like to find that out again. I think that's a critical question."
The Dean of Students Office gets over 100 reports of cheating in class work each year, Reitman said. "I don't know if that's the tip of the iceberg."
