Education Secretary Betsy DeVos holds a news conference at the Department of Education on July 13. (Evelyn Hockstein/For The Washington Post)

Lara Bazelon is an associate professor and director of the criminal-justice clinics at the University of San Francisco School of Law. John Villasenor is a nonresident senior fellow at the Brookings Institution, a professor of public policy and engineering at the University of California at Los Angeles, and a visiting professor at the UCLA School of Law.

This month, Education Secretary Betsy DeVos signaled that her department will likely revise an Obama administration policy on how colleges and universities handle campus sexual assaults.

DeVos, who observed that “a system without due process ultimately serves no one in the end,” is right to address this topic. And she has her work cut out for her. The rhetoric that dominates the debate often obscures the notion of ensuring fundamental fairness — to both sides — when adjudicating a sexual assault accusation.

The Obama administration’s Title IX policy, established in 2011 through a letter issued by the Education Department’s Office for Civil Rights, came on the heels of reports that colleges and universities were giving sexual assault accusations insufficient attention — or ignoring them altogether. The policy requires all colleges and universities that receive federal funding to convene investigative proceedings to adjudicate allegations of sexual violence and sexual harassment made by students.

It is vitally important to provide justice to sexual assault survivors, but in pursuing that goal, the Obama administration’s policy has created extraordinary risks for the subset of accused students who are innocent. Under the 2011 policy, accusations must be evaluated not under the “beyond a reasonable doubt” burden of proof used in criminal courts, but instead under the much lower “preponderance of the evidence” standard, which requires a finding against the accused if the probability of guilt is more than 50 percent. This means that, even if the tribunal reviewing the evidence concludes there is a 49 percent chance that accused students — the overwhelming majority of whom are male — did not engage in the alleged conduct, the accused will nonetheless be found responsible and, in some cases, expelled.

In addition to the low standard of proof, the 2011 policy contains fundamental flaws of due process. It recommends, but does not require, the availability of an appeals process, and it imposes no requirement that appeals be heard by an independent entity. As a result, many campuses manage appeals through the same office that oversaw the initial proceeding, creating a bias in favor of affirming the original finding.

The current approach also fails to guarantee the accused access to a hearing, a right to review evidence against him or the right to ask questions through a lawyer or other appropriately qualified representative. Unsurprisingly, this can lead to the kinds of sham proceedings that have attracted national headlines in recent years.

Actions at the federal level alone may not be enough to address the flaws in this policy. Supporters of the current framework are already devising plans at the state level to compensate for a potential federal rollback. In California, for example, state legislators have drafted legislation that would, in effect, turn the 2011 federal policy into a state law.

This would be a mistake. If accusations of sexual violence are indeed to continue to be adjudicated by on-campus tribunals — and we note that the criminal courts are in many ways much better equipped than campus administrators to manage trials regarding acts that constitute violent felonies — then the 2011 policy requires a major overhaul.

Probability models show that low standards of proof risk unacceptable results: If the “preponderance of the evidence” standard were used in the regular criminal-justice system, innocent defendants would face a chance of false conviction as high as 33 percent. There is no reason to believe the error rate is less in an on-campus proceeding; indeed, given that an accused student has far fewer protections than a criminal defendant, it is likely to be higher.

A higher burden of proof is necessary, and so are more robust procedures to ensure that proceedings are fair, balanced and transparent. Using a standard of “clear and convincing evidence” — a standard less stringent than “beyond a reasonable doubt” but more stringent than “preponderance of the evidence” — would help reduce (though certainly not eliminate) risk to wrongly accused students.

Due process must be the core component of any campus adjudicatory system. Otherwise, on-campus sexual assault proceedings will continue to be rightly challenged as lacking in fairness and legitimacy.