In his State of the Union speech last night, President Trump cited routine sexual abuse of migrant women making the journey north as one reason to support reduced illegal immigration—by building a wall, or by other means.
"Tolerance for illegal immigration is not compassionate—it is cruel," said Trump. "One in three women is sexually assaulted on the long journey north."
This statistic is dubious. As The Washington Post explained, it comes from an unrepresentative sample of just 56 migrant women, 31 percent of whom said they were "sexually abused" during their journey. Just 10 percent said they experienced sexual violence. The study does not clarify the difference between "abuse" and "violence," but suggests that abuse is some lesser category of unpleasantness—nonconsensual sexual contact that does not rise to the level of violence, and trading sex for food, I would guess. In either case, this was a small number of victims.
Another statistic frequently cited by immigration hawks contends that 60 percent of migrant women suffer sexual violence as they make their way to the United States. This figure is even more seriously flawed, and is not based on anything current or reliable. It "comes from interviews conducted a quarter-century ago, making it irrelevant now," writes The Post's Glenn Kessler.
Trump is citing bad data in service of a bad cause, and people should be aware. But while we're on the subject of outdated statistics being used to advance an illiberal policy agenda, I can't help but notice some similarities between misleading claims about sexual assault among migrant women and misleading claims about sexual assault among campus women.
After all, many politicians, activists, and journalists have advanced the idea that as many as one-in-five female students will be victims of sexual assault while on campus—many of them at the hands of serial predators, who commit rape over and over again until they are apprehended. And while The Washington Post has been much more evenhanded and nuanced in its approach to campus rape reporting than other mainstream publications, a 2015 survey co-sponsored by the paper and the Kaiser Family Foundation that offered support for the one-in-five statistic failed to make distinctions between forced assault and nonconsensual touching—precisely the kind of error committed by Trump last night (assault vs. abuse).
Similarly, if we should be skeptical of the 60 percent rate of abuse for migrant women because the figure is 25 years out of date, we should also be skeptical of the data that undergirds the serial predator theory of campus sexual assaults, which was collected at a commuter school—the University of Massachusetts at Boston—during the 1990s.
Under the Obama administration, dubious sexual assault statistics were used to justify new policies that eviscerated due process protections for students accused of sexual misconduct and also imperiled free expression. The new administration, to its credit, is working hard to undo these changes.
Trump is far from the only authority figure to cite bad data in order to guilt-trip the public into supporting his ill-conceived policy ideas. It's important to call out the president, but keep in mind that there are plenty of other "zombie statistics" in our midst.