After Cornell, boards want prevention plans. Can yours prove it works?

The New York Times' almost-daily coverage of Cornell has put one question on every university board's agenda: What is our prevention plan? The harder question comes next: does the plan reduce assaults, and does the school have proof?

Sexual assault is already happening on many campuses, whether or not it has made the news. CrisisCommand® was built for leadership teams in the middle of a crisis, and it is just as useful while campus sexual assault is still an issue on the board agenda and not yet a headline about the school. Many universities have answered the board's question after a crisis of their own. I pulled three of those answers from the campus sexual-assault cases in its Knowledge Base, and they took very different routes.

Three answers to the same question

Dartmouth acted fast and measured activity. In January 2015, under pressure over fraternity culture and falling applications, Dartmouth banned hard alcohol and pledging, required a four-year sexual-violence prevention program and announced six residential communities backed by about $1 million a year in programming. An outside panel reported in December 2016 that every initiative had progressed and saw "some early indication" of less high-risk drinking. The review tracked whether the changes were made. It did not report whether sexual assaults went down.

Harvard used data and penalized the wrong thing. In March 2016, a Harvard task force reported that 47 percent of female seniors involved with the all-male final clubs had experienced nonconsensual sexual contact, against 31 percent of all female seniors. Harvard responded by barring members of single-gender social organizations from leading student groups, captaining teams and receiving the dean's endorsement for top fellowships. Fraternities, sororities and students sued for sex discrimination in 2018, and Harvard stopped enforcing the policy in June 2020. Women's groups bore the heaviest cost: Harvard had nine sororities in 2014 and one by December 2018.

Columbia studied first. After 23 Columbia and Barnard students filed federal complaints in 2014, Columbia put more than $2 million into SHIFT, a research project run by its public-health school. Researchers surveyed about 1,600 undergraduates and followed hundreds more through diaries, interviews and observation. Columbia exempted them from its mandatory-reporting rule, so students could describe what happened without triggering an investigation. SHIFT found that 22 percent of undergraduates had been sexually assaulted since enrolling, more than half while incapacitated, and that only 2.2 percent of them filed a complaint. Columbia then changed specific places and requirements: a dining hall open 22 hours a day so students had somewhere to go late at night besides a dorm room, reconfigured lounges, themed residences and a mandatory prevention program that now reaches more than 12,000 incoming students a year.

Why schools reach for bans

A university under pressure picks the response its critics will see on the day it is announced: a ban or a penalty. Swarthmore ended its fraternities and sororities in May 2019 while students occupied a fraternity house and the president's office; the outside investigation that followed found no direct evidence of policy violations by current students. That decision did not start from evidence about how assaults were happening on Swarthmore's campus.

The sequence repeats. A case breaks, the board asks for a plan, the administration announces the most visible fix, and the progress report counts what was implemented. When the next case breaks, nobody knows whether the number moved.

What the board will be asked next

A university that approves a prevention plan this fall will be asked whether assaults went down. Parents will ask it before they send a child to campus. I am about to send my daughter to college, and I will be one of them. A school whose only evidence is its complaint count has almost no evidence: Columbia found that only 2.2 percent of assaulted students filed a complaint, and in 2024 only 6 percent of those affected reported to its Title IX office.

What works

Survey students confidentially before choosing a fix. Columbia exempted its researchers from mandatory reporting so they could promise students confidentiality, which is how the study captured assaults that never reached the Title IX office.

Remove the reasons students don't report. In 2016, Brigham Young University promised amnesty from Honor Code discipline to students who report an assault and separated its Title IX office from its Honor Code office.

Change the specific places the data points to. Columbia's late-night dining hall and redesigned lounges answered what its research found about where students went late at night.

Start prevention education before students arrive. A 2018 SHIFT study found that students who learned refusal skills before age 18 were less likely to be sexually assaulted in college.

What does not work

Treating complaint counts as the size of the problem. A quiet complaint log measures reporting. At Columbia, reporting captured a small fraction of assaults.

Penalizing a group without testing the policy in law first. Harvard's club sanctions were reversed after a sex-discrimination suit, and the sororities lost the most.

Reporting what was done in place of what changed. Dartmouth's review confirmed every initiative had progressed and did not report whether assaults fell.

Changing the survey every few years. Columbia used a national survey in 2015, its own in 2018 and 2020 and a different one in 2024, so it has no clean before-and-after for its own prevention work.

Show the number

Dartmouth reported progress on everything except assaults. Harvard defended a penalty for four years and dropped it. Columbia studied its own campus first. If your school already has a prevention program, be ready to show evidence that it reduces assaults, or change it until you can. If you are building one, survey your students confidentially to learn how, where and to whom assaults happen, fix the specific places it points to, require prevention education before students arrive, and publish the result on the same survey every year.