A Graphic Display: The Anatomy of a Lie
Jan. 8, 2013
It turns out that I thought too much of Dylan Matthews whose post of a rape incidence graphic I responded to yesterday. Turns out Matthews has a history of statistical malfeasance. He is a newly minted Harvard grad though so he has a few more years until he achieves Matt Yglesias status.
I’ll add a couple of more points to the argument. First, the palatability of the information passed around the blogosphere in the form of infographics is a dangerous thing. It’s the blog equivalent to Krispy Creme donuts. Everyone likes them because they’re just so sweet and easy. A lot of energy – good or bad – is packed into one compact morsel.
So you had that graphic which began at a startup feminist organization called The Enliven Project. Matthews considers it a “reputable org”, yet it pretty much was born yesterday, and it has produced no known policy papers and hasn’t been mentioned by any other media outlets or even other feminists until this graphic hit the web.
A Huffington Post blogger tweeted the graphic. Dylan Matthews threw it up at the widely read WaPo Wonkblog. Feministing snagged it from there, Jezebel is passing it around, and now it’s off to the races. The feminist borg has swallowed it and now it is fact. Someone from Business Insider has even been snookered. This is always how it goes, but now with social media we can look into the gear works in real time.
So let’s go over this. The founder of The Enliven Project is Sarah Beaulieu. Responding to inquiries in her methodology she wrote (this is also what Matthews emailed to me after I asked him for data to back the claim):
- Some reports suggest that only 5-25% of rapes are reported to authorities. Other suggest that close to half are reported. We assumed 10%, which is dramatic, but possible.
- Of the rapes that are reported, approximately 9 are prosecuted.
- Of the prosecuted, 5 result in felony convictions. This is across the board for all felony prosecutions, not just rape.
- Assuming that 2% of reported rapes are false and a 10% reporting rate, the graphic assumes that 2 of 1000 rapes are falsely reported (assuming a rape can’t be falsely reported unless it’s reported in the first place)
These are very many assumptions and cherry-picked stats.
Beaulieu admits to choosing a number that is low enough to be considered a “dramatic” estimate of reported rapes. She then cites the percentage of rapes reported from a 1999 study from the National Center for Policy Analysis. From there, to figure out the percentage of rapes that end up as felony charges, she cites a 2006 DOJ study that focused on defendants in large urban counties.
Beaulieu uses the phrase “some reports suggest” and then throws out a low number which sets the table for the bulk of the analysis. The National Institute of Justice found that 36% of rapes are reported to authorities. RAINN, a victim advocacy group, cites a 46% reporting figure which it gleaned from the very same report originally cited by Beaulieu. Which numbers to believe? At the very least, Beaulieu has chosen the most extreme stats that support her claim at every threshold of the process. And, worse, she has been less than honest about the origins of her statistics.
Beaulieu also considers everything reported in victimization surveys as a rape are felonious acts. This is an important point that isn’t discussed very often. Do feminists and other activists think that, say, a man who has sex with a woman who is too drunk to consent should be charged with a felony? Personally, I don’t think that is always a felony. If you could convince me it is a felony, I don’t think it is worth prison time, but feminists might believe that it is. How much prison time does that kind of rape warrant? How many men who are convicted of rape take pleas down to probation and fines? These are tangible punishments based upon the justice system’s available resources.
Feminist activists and their willing pundit poodles don’t have to delve into those details because their followers don’t demand it, and their digital gotchas don’t have room for such nuance anyway.
For her final trick, Beaulieu runs with a 2% false rape accusation figure. She draws this number from the National Center for the Prosecution of Violence Against Women. Even in their seemingly activist cherry-picking, the group repeatedly says in their paper that reliable estimates of false accusation range from 2% to 8%. Beaulieu conveniently chooses the low end. Even then, these figures represent a floor of proven falsities. It is impossible to know how many false rape claims are not proven to be false.
The graphic also misleads by implying that each rape is committed by a different rapist. It explicitly denotes each stickman as an individual “rapist”. Research on this question has found that the majority of rapes are committed by repeat offenders. Lisak and Miller (citing doesn’t mean that I condone every aspect of their methodology) found that among a sample of over 1,800 men, 120 self-reported actions that constituted rape according to the parameters of the study. Those men committed 483 acts of rape. This figures out to 4.025 acts per offender. Applying that statistic, 248 rapists would have committed those 1000 rape acts in Beaulieu’s graphic. Tweaking the assumptions to a 40% reporting rate and an 8% false claim rate, we would find 32 false rape accusations. Compare that to the actual number of rapists, 248. Two versus 1000 has a much different feel than 32 versus 248.
And from there you start thinking about the decision on the part of the woman who is deciding whether or not to report. Feminists assume that every single act that they define as rape is something that should be reported. Rape has elements (repeat: elements; they aren’t the same) of both theft and assault. But very many people decide not to report either of those acts. Feminists would have it that every act which they define as rape would end up in a man serving time in prison. The truth is that feminists are trying to convince women that they should be more pissed off about having experienced rape. That they should decide to report all rapes including ones which are deemed rape due to lack of explicit consent. But in so many other instances we’re told to “trust women” and their decisions. Why stop now?
A snippet from The Enliven Project’s About page is telling:
An American man’s chance of prostate cancer is exactly the same as his chance of being sexually assaulted: 1 out of 6. Among American women, 1 of 8 will be diagnosed with breast cancer; 1 of 4 will be sexually assaulted. Yet in 2011, the Susan G. Komen Foundation spent $492 million on awareness and prevention, five times more than the four largest anti-sexual violence organizations combined.*
Besides the “1 in 4″ myth which has been disproved, the logic behind this passage is telling. Breast cancer is not a crime. There is much more need for private organizations to donate money to fight the disease. The boundaries are also much clearer. But in the U.S. we do have resources to fight crime, including rape. It is erroneous to think of fighting rape in terms of how much money private organizations donate to the cause. And it would be almost impossible to determine how much taxpayer money goes to directly and indirectly investigating, prosecuting, and thwarting rape via the justice system, police agencies, and within social, family, and community networks.