Matt Walsh Disputes Cornell Survey as Hochul, Letitia James Handle Case

Conservative commentator Matt Walsh says a Cornell University survey finding that 35 percent of undergraduate women experienced nonconsensual sexual contact cannot be trusted, as Gov. Kathy Hochul and Attorney General Letitia James face criticism over their handling of the Cornell 7 case. His reason is that only a small share of the students invited to take the survey answered it.

The figure has become a talking point in the fight over the so-called Cornell 7 case. In that case, a former student has sued seven fraternity members, saying they gang-raped her at a fraternity house in October 2024. New York Governor Kathy Hochul has since named Attorney General Letitia James as special prosecutor, a choice that has drawn criticism from the right. Walsh devoted an episode of The Matt Walsh Show to the case.

"If only 15% of undergraduates reply to your survey and 35% of the respondents say that they were sexually assaulted, that obviously does not mean that 35% of undergraduates say they were assaulted," Walsh said.

"The people who are going to respond to a survey like this are far more likely to have a complaint," he added.

What the Cornell Survey Actually Found

The 35 percent figure comes from Cornell's 2025 Survey of Sexual Assault and Related Misconduct, which was carried out in spring 2025 and published in October 2025. ABC News reported the figure last Wednesday, and Walsh played that report on his show.

The survey went to a stratified random sample of 6,000 students across the Ithaca, Weill Cornell Medicine and Cornell Tech campuses, and 926 responded. That is an overall response rate of 15.4 percent, down from 36 percent in 2023 and 38 percent in 2021. Among undergraduates the rate was 12.8 percent, not the 15 percent Walsh cited. Walsh read the report's own line about undergraduates responding "at a lower rate of around 13%" on air just before he made that claim.

On the main point, the report agrees with Walsh. It says the low share of completed responses, "coupled with response patterns suggesting hesitancy to disclose detailed information about incidents, limits confidence in the accuracy of these estimates." It adds that the 2025 estimates "are best interpreted with greater caution than those from prior administrations."

Cornell said in its statement on the results that the lower response rate "limits the reliability of estimates."

The Definitions Walsh Targeted

Walsh called the 35 percent number "fake too." He pointed to the survey's four tactics: physical force, incapacitation due to drugs or alcohol, coercion and the absence of affirmative consent. He argued that this broad framing inflated the count.

But the 35 percent figure counts only two of those tactics: force or threat of force, and incapacitation. Contact involving coercion or the absence of affirmative consent is not part of it. The figure covers penetration, sexual touching and, for the first time in 2025, unwanted choking or strangulation. It counts any such experience since the student entered Cornell, not just the past year.

The rise is also sharper than past results. On the same measure, undergraduate women's reported rate was 20 percent in 2021 and 23 percent in 2023. In the 2019 Association of American Universities campus climate survey, which covered 33 schools, 25.9 percent of undergraduate women reported such contact since enrolling. That survey's response rate was 21.9 percent.

A different federal measure gives a much lower number. The Justice Department's National Crime Victimization Survey found a rate of 6.1 rapes or sexual assaults per 1,000 female college students each year from 1995 to 2013. That survey counts incidents within a single year and uses crime-focused questions, so the two measures are not directly comparable.

Refugee Camp Comparisons Mix Different Measures

Walsh also compared the Cornell figure with conflict zones. He cited a study of a camp in Mogadishu, Somalia, in which 19 percent of women reported gender-based violence in the previous 12 months. He also cited Sierra Leone's civil war, where he said 9 percent of women reported sexual violence.

"We're supposed to believe they're being assaulted at triple the rate of women in war-torn African hellholes," Walsh said.

Those numbers measure different things. By Walsh's own description, the Mogadishu figure covers a single year and all forms of violence, while Cornell's covers a student's entire time on campus and a defined set of sexual contact. A 2002 population-based assessment by Physicians for Human Rights, with UNAMSIL support, examined war-related sexual violence among displaced people in Sierra Leone. That is a narrower category than the broader set of experiences Cornell's survey counts.

What Happens Next

Walsh used the survey to argue that "women are being trained to recast unpleasant sexual experiences as assault." He said the case shows that consent-based sexual ethics have failed.

Cornell has said it will hire an outside law firm to review how the university handled the accuser's allegations. ABC News reported that James' office is reaching out to Jane Doe as part of the reopened criminal investigation. The university is legally required to run the survey every two years, so the next round will show whether a higher response rate confirms or reverses the 2025 jump.