Skip to main content

Decoding Honor

What 99 years of data say about the Honor System

Sitting in the second row of Monroe 120 during an economics midterm, third-year student Gray had a choice. He could leave the question incomplete or, with the slightest shift of gaze, lift the answer from the paper of the classmate beside him. Gray picked Option B, leaving for later the matter of his pounding heart and sweaty palms when he turned in his work.

Later came sooner. The professor had given adjacent test takers different data sets to use in solving the same problem. Gray’s answer was a perfect match with his neighbor’s and therefore exactly wrong.

The call came about a week later. An Honor System officer informed Gray he had seven days to admit to cheating or undergo Honor’s adjudicative process. After much personal torment, Gray confessed and left the University.

He tells the story more than a year later over lunch on the Corner. He’s back in town because of Informed Retraction, the multistep Honor System mechanism that lets offenders resume their studies after a two-semester suspension. Gray is not his real name. It’s a pseudonym suited to the Honor System’s now-established middle ground between the black-and-white choice of either full exoneration or permanent expulsion. Completing the metaphor of Honor’s move to a new era of forgiveness, after lunch the returning student hops on a Lime electric scooter so he’s not late to his next class.

Gray isn’t just a human story; he’s a statistic. His case adds to the 61 Informed Retractions tracked in the newly released Honor Bicentennial Report, the most comprehensive data analysis of the University of Virginia Honor System ever undertaken. Honor Committee volunteers last summer sifted through nearly 100 years of Honor data, reconciling the less-than-uniform record-keeping of about as many honor committees. Graduating third-year student Charlotte McClintock (Col ’19) served as principal author, and the report’s analysis is as eye-opening as it is empirical.

This graph shows a century of Honor dismissals (expulsions, voluntary exits and, after 2013, one-year suspensions), save for data gaps during World War II, 1972 and 1985–86. A physics professor’s computerized crackdown on cheating caused the 2001 spike. Source: Honor Bicentennial Report, Feb. 2019

Suspension System

The report offers the first objective evidence of Informed Retraction’s impact, and it’s dramatic. Voted in in 2013, IR allows an accused first offender to confess within seven days, follow guidelines for making amends, and take a two-semester leave. The other choices: claim a contributing mental disorder, which is a complicated process, or gut out an Honor hearing where, short of acquittal, the single punishment is expulsion. Says Gregory “Ory” Streeter (Med ’19), the immediate-past Honor chair who spearheaded the study, “This other path … really defines our system now.”

IR’s most noticeable effect has been a dramatic drop in the number of Honor hearings. In 2012, the year before IR entered the system, there were 29 proceedings. By 2015, three years into IR, the Honor docket had shrunk to 12, and then seven in 2016 and eight in 2017, the last year of the data set.

IR has become so established within the Honor System that, over those three years, 40 percent of accused students opted for it. It has made for faster case resolutions: More reckonings take place at the front end of the Honor System, within those first seven days of accusation, rather than months or semesters later at trial.

With the new efficiency has come a higher rate of punishment. In 2012, 43 percent of confronted students received sanction, leaving the University for good, either by volition or verdict. By 2017, with IR added to the mix, the sanction rate surpassed 50 percent, with suspensions outnumbering expulsions by 4-to-1.

Darden School of Business professor Michael Lenox (Engr ’93, ’94), a former Honor chair, sees Informed Retraction as a return to Honor’s roots, when accusations were resolved not by the pounding of a gavel but a knock at the door. “In some ways, it’s getting, maybe, back to that old spirit of, you did not live up to the standards, there’s a conversation, there’s a consequence.”

The difference is that instead of permanent banishment, IR accommodates rehabilitation and return. Because of that alignment with contemporary sensibilities, along with the swifter justice, Streeter says, “The Honor System is a stronger system today than it has ever been before.”

The Demographic Over-Under

The Honor Bicentennial Report offers the most penetrating insights into Honor’s demographic factors, though they come with caveats. The annual case numbers have always been small—it averaged 46 cases or “reports” in the six most recent years—so the study is careful not to overstate patterns. The oldest data are the least detailed, providing numbers of students sanctioned but not the total number charged, let alone identifying characteristics about the parties or, often, even the nature of the offense. Demographic characteristics begin to come into focus in 1987, with the data from 2012-17 offering the most specific breakdowns. Even those statistics have qualifiers, including sizable numbers of “race unknown,” particularly for international students.

Where trends do emerge, the report carefully refers to a particular group as being “over-represented” or “under-represented” relative to their proportion of the overall student population. It identifies significant deviations without identifying causes or recommending solutions, leaving that to future initiatives. Among the findings:

  • African American students have been over-represented and whites under-represented in Honor enforcement, but the disparity between black and white has narrowed over the last 30 years. From 1987 to 1989, African Americans made up 9 percent of the student population but 42 percent of those sanctioned for Honor offenses. A couple decades later, from 2010 to 2016, they comprised about 6 percent of the population and 12 percent of those punished. In the past three years, their sanction rate dropped to 3.6 percent, according to McClintock, the study’s author.
  • Asian and Asian American students are the most over-represented race in the sanction data. In 2015, they constituted 10 percent of the student population but, from 2014 to 2016, 53 percent of students sanctioned.
  • Hispanic students have few Honor troubles relative to their size of the student population. From 2012 to 2017 they represented 6 percent of students but drew only 3.6 percent of Honor penalties.
  • Men continue to have more Honor trouble than women, but women have been catching up. From 1990 to 2000, with the male-female population split staying within 3 percentage points of 50-50, men commanded 70 percent of punishments. By 2014 to 2016, UVA was roughly 46-54 percent male-female, and the men incurred 49 percent of the penalties.
  • The greatest disparity concerns UVA’s increasing numbers of international students, who accounted for 17 percent of the population from 2014 to 2016 but 55 percent of sanctions.

Streeter says, “We have to educate ourselves out of that discrepancy. We’re either not preparing faculty to understand their [own] biases, or we’re not preparing faculty to meet the needs of our international students, or we’re not preparing our international students to meet the demands of the University education here.”

He sees a familiar historical pattern. Each time the University becomes more diverse, the new cohort encounters disproportional Honor trouble, whether through heightened scrutiny or individual factors. That happened with the influx of Northerners from 1919 to 1939, when out-of-staters, especially New Yorkers, drew 67 percent of Honor sanctions. It happened with African Americans, starting in the 1960s. Women were the exception, though they too faced skeptics who fretted that coeducation in 1970 would strain the Honor System.

“We’re sort of a canary in the coal mine in terms of the University academic experience,” Streeter says. “But time and time again, the University has risen to meet the challenge of supporting these newest members of our student body. And our international students are in the same place.”

Race is known for only 36 percent of international students. Under federal guidelines, they’re counted as “non-resident alien.”

Big picture: It’s small

In its grand sweep, the omnibus report shows Honor’s narrowing scope. The system has largely defaulted to a mechanism for policing academic misdeeds. From 2005 to 2016, cheating cases took up 67 percent of the punishments meted out. It stands to reason: Professors and teaching assistants brought 73 percent of the cases from 2012 to 2017, students reporting just 18 percent.

Stealing cases disappeared during 2014 to 2016, continuing at least a decade of decline, perhaps a sign that the University Judiciary Committee, with overlapping jurisdiction and more forgiving penalties, has become theft’s forum of choice.

Then there’s Honor’s perennially low case volume. It’s one area where Informed Retraction has had little influence, despite its promise of encouraging more witnesses to come forward. There were 42 case reports in 2012, the year before IR came into effect. The number spiked to 56 in 2014, but has declined each year since in the study, dropping to 39 in 2017.

It’s tempting to blame the repeal 40 years ago of the provision that made it an offense to witness an Honor crime and not report it. In combing through all the historical data, however, the team found no instance of the so-called “non-toleration clause” ever having been enforced.

The Honor System addresses only a tiny fraction of actionable conduct. In an anonymous survey several years ago, 4.7 percent of UVA students admitted having committed an offense. (How that compares with earlier generations is unknown; they were never asked.) With only 0.2 percent of students facing charges in recent years, the resulting math suggests that some 96 percent of violators never have to answer for their Honor crimes.

But zeroing in on case volume may miss a larger point. Compare UVA’s low single-digit rate of self-reported transgressions, for example, to the 68 percent of undergraduates surveyed nationwide who admitted to academic cheating. Says Lenox, the Darden professor and Honor observer, “Wouldn’t a city like to have lower crime, versus more arrests?”