IN A FEAT of irony, the Honor Committee has been either actively dishonest or incompetent. Neither interpretation is flattering to the Committee.
Last year, before students considered the four honor referenda, three of which the student body soundly rejected, the Committee in conjunction with the Office of Institutional Assessment conducted a survey to gauge student opinion on the honor system. The Committee concluded in a press release: "Most importantly, though, only 6.5 percent of respondents believed they had committed an Honor offense. This figure can be compared to a study by Donald L. McCabe that discovered that 75 percent of students at large public colleges and universities have cheated at least once ("Toward a Culture of Academic Integrity," Chronicle of Higher Education, Oct. 15, 1999)." This is supposed to suggest that cheating at the University is considerably less prevalent than at other institutions.
Unfortunately, the Committee's research methods are so bad at calculating rates of cheating at the University that their conclusion is disingenuous and misleading. The irony here is obvious: Disingenuous and misleading research is what the Committee ostensibly opposes.
The survey was sent to 2,100 students. First and fourth years, racial minorities and graduate students were oversampled according to Chairman Thomas Hall in an e-mail response.
|
The survey got over 1,600 responses. All of the respondents elected to respond to the survey on their own. This self-selection renders the survey unscientific at the outset. When researchers are incapable of controlling who responds, they can't know what the probability each member of the population has of getting into the sample. In short, it is as unreliable as a People magazine online poll that elected Hank the Angry Drunken Dwarf the most beautiful person by over 200,000 votes.
The Committee's study is the equivalent of a modern straw poll, from which absolutely nothing meaningful can be inferred. It makes sense that those who are most passionate about the honor system are those who would make the effort to fill out the survey online. And let me assure you, plenty of effort is involved. The survey is more than five printed pages in length. If those who feel passionate about the honor system are overrepresented in the sample, which they most likely are, it is not surprising that the Committee would find that there is a relatively low incidence of cheating among students. Political scientists Robert Erikson and Kent Tedin write of this sort of research: "We know that those who 'self-select' ... tend to [respond] to those to whom they feel are sympathetic."
Hall says: "This was the most professional survey that can be administered by the University." That's pretty sad. If what Hall says is true, the University should remain out of the business of conducting surveys.
Setting aside these methodological objections, the inferences the Committee wishes to make remain flawed. For one, what incentive is there for any respondent, even on an anonymous survey, to admit that he or she has cheated? Even if there are those who are feeling charitably honest on an anonymous survey, there are still those who aren't. That means that the results would logically report fewer cheaters than there actually are. The Committee found only 6.5 percent of students that admit to cheating, but this problem, compounded by the sample biases, render the conclusions suspect. The Committee doesn't mention in its press release that the survey found over 13 percent of students report that they "maybe" committed an honor offense. The honor system is so complex now that students don't even know when they've cheated, which clearly could contribute to underreporting the presence of cheating.
But perhaps the strongest evidence against the Committee comes from the Committee itself. In the same press release dated Feb. 23, 2001, the Committee reports: "The survey revealed strong support for the upcoming referenda on the Honor Committee Constitution." Here's a news flash. Students rejected three of four proposals, proving this survey a certifiable deviation from reality. One is compelled to wonder if the Committee constructed the survey to make the proposals seem desirable and the incidence of cheating small. In any case, the Committee wasn't being scientific.
Which brings us to the unflattering dilemma. Either the Committee did poor research on purpose or it didn't. If it didn't, the Committee is merely incompetent. That's unsettling. If the Committee is aware of its bad research and the bad inferences drawn from it, then the Committee has been dishonest. That's even more unsettling.
(Jeffrey Eisenberg is a Cavalier Daily associate editor. He can be reached at jeisenberg@cavalierdaily.com.)
Read More
EDITORIAL: Collectives are a misinterpretation of the mission of NIL
By Editorial Board | YesterdayA large driver of this inequality comes in the form of collectives, the structures many schools use to organize their NIL payments.
CARTER: Data science without sustainability is insufficient
By Nathaniel Carter | 4 days agoThe University appears committed to a future defined by data, but lacks the dedication to meeting this future’s energy needs in a sustainable way.
SULLIVAN: Gov. Youngkin purges integrity from voter registration process
By Scarlett Sullivan | 6 days agoYoungkin once again abandoned his duty to protect Virginian voters’ rights, setting a dangerous precedent for elections.
