How the Media and You Are Misled by False Data

How the Media and You Are Misled by False Data

  • Statistics get repeated, even though they’re false
  • Lawmakers use bad data to make policy
  • Americans live with the consequences of failed policies
The Fiscal Times

The most obfuscating and misleading arguments made in debate of any kind usually begin with the words, “Studies say ….” People passionately arguing for a favored position will resort to these citations of assumed authority, and will often fail to comprehend the scope or underlying data of such published material. Even on line, where writers usually link to the source data, the studies either prove to be limited in application, poorly researched, or entirely wrong.

Sometimes that has serious consequences. A study published in the British medical journal Lancet more than a decade ago started a panic about a supposed causal connection between vaccinations and autism. It fueled an anti-vaccination movement that has resulted in the return of diseases once thought stamped out in the West such as measles, whooping cough, and more.

Related: The Closing of the Millennial Mind on Campus

The study was later exposed as a fraud, based on only twelve subjects handpicked by its author to support his conclusion, with the data even further manipulated. The Lancet later withdrew the study and admitted it was “an elaborate fraud.” By then, it was far too late to undo the damage done to uncounted children over several years.

Most questionable studies, and questionable claims made from them involve less malice and intent to defraud but matter nonetheless for public policy. Other times, anecdotal data is used to eclipse more comprehensive information.

Faulty claim # 1: We saw some of this in the debate over the HHS contraception mandate, when policymakers arguing in favor of the mandate argued that women were being broadly denied access to contraception by schools and employers. They used anecdotal and narrowly based data in support of that point.

That ignored a much broader study published by the CDC in 2009 that entirely negated that argument. In a 26-year study of unwanted pregnancies – the very issue that HHS proposed to reduce through the mandate – the CDC found no evidence of a lack of access to birth control as a driver of unwanted pregnancies. “Contraceptive use in the United States is virtually universal among women of reproductive age,” the study noted, and “99 percent of all women who had ever had intercourse had used at least one contraceptive method in their lifetime.”

Related: Hobby Lobby Ruling Affirms America’s Independence

Readers who understandably consider the citation of a study here ironic are welcome to follow the link and peruse the data for themselves. The only mention of access as a problem to be solved is access to health insurance in a footnote on page 42. It’s worth noting too that this is the government’s own data, which was largely ignored by the White House while portraying access to contraception as a crisis in the United States.

Faulty claim #2: That brings us to another crisis declared by the Obama administration – the alleged epidemic of sexual assaults on college campuses, and the existence of “rape culture” in higher education. From Barack Obama and Joe Biden on down, the White House has tossed around statistics supposedly from reliable studies of the issue, claiming that 1 in 5 women in college will be the victim of sexual assault. They have used those statistics to pressure colleges into abusing due process in Title IX enforcement actions, with horrendous outcomes for those falsely accused.

Every rape is a heinous crime, and every victim deserves justice. However, the numbers thrown out to justify this as a specific and acute crisis make no sense at all. Once again, the government’s own data rebuts the claims of a rising epidemic. The Department of Justice reported in December  that the incidence of sexual assaults on female college students between 1995 and 2013 was 6.1 cases for every 1,000 students or 0.61 percent, which was lower than for non-students in the US (7.6 per 1,000). Both are orders of magnitude lower than the 20 percent figure used by the Obama administration. Half of the sexual assaults for college students took place away from where they lived, with presumably some significant part of those off-campus.

Related: The Bias Behind the Rolling Stone Rape Travesty

So where did the White House get its evidence of a “rape culture” epidemic? Unsurprisingly, from another study that has now been seriously challenged. Reason’s Linda LeFauve found that a study by University of Massachusetts-Boston Professor David Lisak, used by activists to argue for the existence of a crisis dishonestly claimed the research of students as his own, and the data had “no direct connection to campus sexual assault.”

The data from which Lisak derived his information was intended to research childhood sexual abuse experiences among adults. Only five questions dealt with potential criminal behavior among the sample as adults, and did not specify where the assaults took place.

LeFauve attempted to find out how Lisak concluded that these were actual perpetrators of sexual assault and whether those assaults took place on campus. Of the 1,882 respondents, only 120 qualified as potential offenders and only 76 as potential multiple offenders – a key point in the claim that universities tolerate a “rape culture.” Lisak claimed to have interviewed “most” of them personally, as LeFauve reports.

Related: Guilt by Association—How Colleges Deal with Campus Sex

“That was a surprising claim, given the conditions of the survey and the fact that he was looking at the data produced long after his students had completed those dissertations,” LeFauve writes, “nor were there plausible circumstances under which a faculty member supervising a dissertation would interact directly with subjects. When I asked how he was able to speak with men participating in an anonymous survey for research he was not conducting, he ended the phone call.”

To recap: in a non-random sample from surveys on child abuse in the male population of UM-Boston employees, Lisak found 4 percent who might have been multiple offenders for sexual assault, and refuses to discuss how he verified that data. This study has since informed the White House, the documentarians behind The Hunting Ground, and other advocates for short-circuiting due process as a response to an “epidemic.”

The Washington Examiner’s Ashe Schow points out that Lisak’s data has been specifically used to justify that rush to condemnation. “If one believes that 20 percent of college women will be raped by a small number of serial rapists, and that very few of those women are lying,” she concludes, “then there's a good chance that each new accused student is a serial rapist-in-waiting.”

In other words, we have the perfect brew for yet another moral panic to cloud our collective judgment. Due process exists to protect people from mob rule and moral panics, as well as to protect us from those who would stoke those panics for their own political purposes. Claims of support from “studies” for extraordinary and yet oh-so-convenient claims need much more careful scrutiny – and perhaps much more pointed skepticism.