Toxic incentive structures surrounding publishing scholarly work may be eroding faculty and research integrity at large. Without proper mediation, those shortcuts can resurface years later at the highest echelons of academia.
At least three former or current college presidents have been accused of some form of academic malpractice in the past 15 months, including Marc Tessier-Lavigne of Stanford, Claudine Gay of Harvard and, most recently, Darryl J. Pines of the University of Maryland. Tessier-Lavigne and Gay have since resigned from their posts. Pines has requested an independent investigation following widespread scrutiny, The Baltimore Banner reports.
Other notable examples include Alade Mcken and LaVar Charleston, the DEI officers of Columbia and the University of Wisconsin-Madison, respectively. Furthermore, work by renowned Harvard behavioral scientist Francesca Gino has been retracted and revised and is under further investigation for plagiarism and data manipulation, Science reports.
“It’s really hard to know what happened in a lot of these cases,” says Julia Strand, professor and chair of Carleton College’s psychology department. “What’s interesting about these high-profile cases is that [it’s affecting people] who have established careers.”
Leaders accused of cheating will exacerbate higher education’s political divisiveness, says Phillip Magness, senior fellow at the Independent Institute, a think tank. “The only people that seem to still have a high opinion of higher ed are on the political left. Part of that is due to the barrage of scandals.”
How to human-error-proof your research
Significant portions of a 2002 scholarly paper co-written by President Pines closely replicate that of an Australian university student six years prior, with up to one-third of the paper suspected to be taken verbatim, according to The New York Times.
Strand, however, isn’t quick to believe plagiarism is the result of malicious action, given how easy it can be to detect. “The line between incompetence, sloppiness and malfeasance is not always clear,” she says, “but the solution for almost all of these things is.”
The University Business Podcast: Why HSIs play a vital role for the U.S. economy
By studying safety culture across construction, mining, medical care and other industries, Strand has developed a series of recommendations for academics interested in human-error-proofing their research systems and workflow.
Some of her tips include:
- Implement error prevention techniques from other disciplines: “[Study] industries in which humans making mistakes kills someone or costs a lot of money.”
- Decide on your research and analysis plan, and submit it to a public registry: Known as pre-registering, this technique can prevent researchers from making post hoc changes during analysis, like p-hacking, cherry picking or hypothesizing after results are known
- After publication, publish your work in a publicly accessible repository: The Open Science Framework helps different communities collaborate on your work and find errors missed during review.
Staying accountable
Unfortunately, not all forms of cheating can be blamed on human error. Higher education officials must contend with cheating from their students as early as the college application process. Fears surrounding AI have only exacerbated the problem.
As a result, higher education leaders must increase scrutiny and consequences for academic misconduct at all levels to nip bad habits in the bud. “Cheating is fairly widespread in universities,” Magness says. “Students carry it with them into professorships and places of their careers where they should absolutely know better.”
And leaders must remain particularly stringent with holding faculty accountable when they’ve been found to deliberately cheat, Magness says. “You can’t just have a plagiarism policy or a cheating policy on paper that you only apply to undergraduates in the classroom.”
Magness also encourages leaders to promote viewpoint diversity to improve the rigor of their peer-review processes. “If universities [only have faculty] almost entirely composed of one side, that’s the only viewpoint that will be represented in peer review. There’s no longer an incentive to scrutinize something that you already agreed with.”
Correction: An earlier copy of this article implied that former Stanford president Marc Tessier-Lavigne had been accused of plagiarism at the time of his resignation. Rather, papers of which Tessier-Lavigne was the principal author were found to contain data manipulation of which he was unaware.
