Math is Hard: Fear is Not Danger

A recent Gallup poll names the Houston, Sugar Land, and Baytown region among the least safe U.S. metro areas, according to resident confidence in the safety of where they live.

Only 63 percent of those polled in the Houston area responded that they felt safe walking alone at night in the area they reside.

* * * * *

Compare that to the 80 percent in the Minneapolis-St. Paul area who feel totally secure walking after dark. Texans might scoff that with frigid Minnesota temperatures, criminals would be stymied to commit violent acts in five layers of clothing. Houston must have crime-friendly weather, for the most part.

(Houston Chronicle.)

Maybe. Or Minneapolitans might more realistically scoff that Houstonians are more frightened than their crime rate merits.

The headline on the Chronicle article is Poll of residents puts Houston on list of least-safe U.S. cities. “Most-frightened” would have been more accurate: there is no strong correlation between violent-crime rates and residents’ fear.

Crime Rates By City

According to FBI Uniform Crime Reporting Statistics for 2010*, Houston’s violent-crime rate (as reported by HPD) edges Minneapolis’s (as reported by MPD) only narrowly: 1071.3 violent crimes per 100,000, compared to 1062.3 for Minneapolis. (So much for the “too cold in Minnesota to jack people” theory.)

Jacksonville residents are as scared as Houston residents, with a violent-crime rate one-third lower (664.4).

Residents of San Bernardino, with a violent-crime rate of 773, are more frightened (61% “yes, safe”) than either Houstonians or Jacksonville residents (63%).

New Orleans has an even lower violent-crime rate (754.4) and more frightened (59%) residents.

Memphis residents’ fear is more in line with their danger. In line, but not necessarily proportional. Whereas out of 100,000 Memphis residents 1,607.8 will be victims of violent crime in a given year, fifty times as many don’t feel safe walking alone at night.***

“Math is hard” is my shorthand for Americans’ tendency to treat fear as risk, and this Chronicle article typifies the problem. Houston is one of America’s least safe cities (in the top third of the biggest 50, albeit safer than Tulsa, Nashville, or Indy, among others); there are solid statistics to put it there. A Gallup poll showing that Texans aren’t as brave as they like to pretend doesn’t show that Houston is unsafe; it shows that they’re unbrave.

*UCR stats are by police department rather than by metropolitan area. I’m using the major city in the area as a proxy for the entire area. Crime rates are generally lower in the suburbs, so metropolitan-area crime rates are generally lower than city-only crime rates. The image is of the fifty largest cities, sorted by violent crime rate.

**Neither Detroit nor Chicago residents appear to have been included in the survey.

***The fear of victimization is partly self-fulfilling. The more people are afraid to be outside at night, the fewer people are outside at night and the greater the likelihood that any of them will be victims of violent crime.

The “Gun-Show” “Loophole” and Tort Law

The “gun-show loophole” is the popular name for the idea that people who wouldn’t be able to buy guns from federally licensed firearms dealers can go to gun shows and buy guns without background checks from non-FFLs.

I’ve been to several gunshows, and haven’t seen more than a handful of guns for sale by non-FFLs. The loophole—if it is a loophole—is, more accurately, a private-transfer loophole.

“Loophole” is a value-laden word; the rule allowing private citizens to transfer guns to each other without background checks is more accurately an exception—an exception that allows gifts between family members as well as sales between strangers.

How significant is the private-sale exception? The New York Times asserts that “Nearly 40 percent of all gun sales are exempt from the system.” I find that surprising, and would have to see the source of the statistics. Any interstate gun sale, even from one non-FFL to another, must pass through an FFL (and so requires a background check). I hunch that intrastate individual-to-individual sales don’t make up 10% of gun sales.

Why does it matter how big the private-sale exception is? Because any new rule has hidden costs, including opportunity costs. When we are rational, we don’t do “whatever it takes” to address our fears; instead we evaluate the costs and benefits of a new rule. For a new rule to be justified, the benefits have to outweigh the costs. If we are merely assuming benefits, we shouldn’t make the rule (even if the costs are also a hunch). The use of misleading labels suggests that there is no sound basis to the rhetoric.

So show me the numbers.

If the gun-show exception is not significant or of small significance, perhaps there is no problem, or maybe there are other solutions to the problem than the use of state violence—for example, tort law. Right now if I had a gun and I wanted to sell it I couldn’t do a background check through NICS—only FFLs have access to the system. Give gun sellers access to the system, and most of them will use it. If they don’t, and if they guns they sell wind up being used to commit mayhem, let the plantiffs’ lawyers at ‘em to hold them responsible for their negligence. It won’t take many kajillion-dollar verdicts to make background checks the de facto rule in private gun transfers.

A Lie Told Often Enough…

Returning to the topic of the Enliven Project rape infographic:

Two Percent?

Sarah Beaulieu, trying to justify the infographic, assumes that only 2% of rape reports are false. This is, as it turns out, not only her assertion; it is also a meme.

Edward Greer of Brookline Massachusetts did some detective work to try to track down the meme's Patient Zero. In an article (PDF, via dinky_hawker on reddit) published at 33 Loyola of Los Angeles Law Review 947 (2000) Greer tracks the percentage (which is by "overwhelming consensus" recognized to be "empirical fact") back to "an illusion that sprang from a mimeoed handout in Susan Brownmiller’s file." Brownmiller's assertion that no more than 2% of reported rapes are false is based entirely on the remarks of a judge, New York Appellate Division Justice Lawrence H. Cooke, "Before the Association of the Bar of the City of New York, Jan. 16, 1974."

Every "scholar" for the last thirty-eight years who has asserted that only 2% of rape reports are false bases this assertion on comment a judge made in a speech to some lawyers. These scholars cite to each other, but in the end nobody has any other basis for the assertion than Judge Cooke's comment.

Twenty-Five Percent?

(This is for that guy at reddit who thinks that saying, "I've got a bias, so consider your own experience"  is less awesome than trying to prove it to him.)

In my original post I mentioned the number "close to 25%"—the exclusion rate in cases submitted to DNA laboratories. Max Kennerly took me to task in comments for not providing a better citation for what I describe as "a credible reason to assume" that the fraction of rape allegations that are false is greater than two percent.

Max, "A credible reason to assume" is auxetic. But, just for you, here's a source: Convicted by Juries, Exonerated by Science, US Department of Justice Office of Justice Programs 1996 (at page 20):

In about 23 percent of the 21,621 cases, DNA test results excluded suspects, according to respondents.

Twenty-three percent of DNA tests excluding suspects does not mean that 23% of rape allegations are false. Many rape allegations don't involve testable DNA; in those that do, suspects may be developed by other means than the complainants' accusation. If a woman is raped and says "I don't know who did it" and the police develop a suspect and that suspect is excluded by DNA testing, that's not a false allegation.

So I'm not sure that the 23% exclusion rate really says anything about how many rape allegations are false. But you can certainly use it as a credible basis to assume. Auxesis.

Forty-One Percent?

There have in fact been studies done of the percentage of rape allegations that are false.

Here are the parameters of one US study, published in the peer-reviewed Archives of Sexual Behavior:

  • 109 forcible rape allegations.
  • Over 9 years (1978-1987).
  • In a midwestern town of 70,000.
  • Investigation of all rape complaints involved a serious offer to polygraph the complainants and the suspects.
  • "False allegation" means that the complainant recanted.
  • The police agency charged recanting complainants with filing a false complaint.

Here are some of the findings:

  • Forty-five of the 109 complainants (41%) recanted, and did not then try to retract their recantations when they were told they would be prosecuted for filing a false complaint.
  • Twenty-seven of the forty-five had made false allegations "to provide a plausible explanation for some suddenly foreseen, unfortunate consequence of a consensual encounter, usually sexual, with a male acquaintance." (I referred to this most common motivation in my first post on the topic; 
  • Twelve of the forty-five had made false allegations to retaliate against a rejecting male.
  • Eight of the forty-five had made false allegations to get attention or sympathy.
  • (Motives were determined based on the complainants' own words; bviously, there is some overlap among motives.)
(Kanin, False Rape Allegations, 23 Archives of Sexual Behavior 81 (1994) (PDF)).

Kanin's data are poo-pooed in an article (PDF) from a magazine published by the American Prosecutors Research Institute’s National Center for the Prosecution of Violence Against Women (bias alert!)—the article on which Beaulieu bases her 2% assumption:

In other words, there is no way to explore whether the classification of these cases as false was simply made as a result of the detectives’ own perceptions and biases, without any real investigation being conducted.This concern is compounded by the fact that the practice of this particular police department was to make a “serious offer to polygraph” all rape complainants and sus- pects (Kanin, 1994, p. 82). In fact, this practice “has been rejected and, in many cases, outlawed because of its intimidating impact on victims” (Lisak, 2007, p. 6).The reason is because many victims will recant when faced with apparent skepticism on the part of the investigator and the intimidating prospect of having to take a polygraph examination.Yet such a recantation does not necessarily mean that the original report was false.

In reality, there is no way that an investiga- tor can make an appropriate determination about the legitimacy of a sexual assault report when no real investigation has been conducted—and the victim is intimidated by the department’s policy of making a “serious offer to polygraph” all rape complainants. As we will discuss at length below, the determination that a report is false can only be made on the basis of findings from a thorough, evidence-based investigation.

The prosecutors' article then goes on to talk about "the MAD study," which classified seven percent of rape reports as false (two of the three authors of the study are directors of EVAW international, which conducted the study; there seems to be no reference to published, much less peer-reviewed results of the MAD study.

People with an agenda can not only pick and choose the statistics they use to make their arguments and info graphics, but also design their studies to get results closer to their ideology. That is one reason that (peer-reviewed and published) > not (peer-reviewed and published).


I'll lay my ideology on the table. I am a criminal-defense lawyer with a mother, a father, a brother, a wife, a daughter, and a son.

I believe that rapists should be reported and prosecuted (I benefit from it not only as a member of society, but also as a professional in the business of defending against rape allegations).

I also believe that false rape allegations should never be made (I benefit from false rape allegations, because they need to be defended against as well, but this (like the War on Drugs) is one of those instances in which I will work against my own self-interest).

"One" would be an unacceptably high number either of false rape allegations or of victims not reporting their rapes. I would no more want one of the men in my family to be falsely accused of rape than I would want one of the women in my family to be raped. I don't know how we can, as a society, reduce the number of unreported rapes without increasing the number of false reports.

I try not to let my ideology affect my presentation of facts here. (That's what the courtroom is for.) "Two percent" is a number based entirely in ideology. Greer, in his article, discusses the ideology behind "two percent":

LDF literature advances the proposition that “women don’t lie about rape” as an axiomatic substrate to their proposed policy changes fueled by the purported two percent false claim figure. As further justification, LDF proclaims that women are deterred from making false rape charges because, inter alia, rape complainants are subjected to a harrowing “second rape.” Simultaneously, LDF wants alterations in the processing of rape charges by reducing the sanctions, costs and trauma—i.e., the “second rape”—that face women who come forward and press rape charges.

Making it easier to press rape charges, Greer points out, makes it easier to press false rape charges. Can we solve one problem without exacerbating the other? I don't know the answer, but I know it won't be found in lies.

Lies, Damn Lies, and Infographics [Updated With Links]

You may have seen this graphic:

…which purports to show that for every 998 rapists, there are 100 rapes reported, thirty people tried, ten people jailed, and only two people falsely accused.

The numbers, particularly the last one—only two false accusations for every 100 true reports—are very interesting to me. Where did the illustrator get them? According to Sarah Beaulieu, who published it, "Statistics from Justice Department, National Crime Victimization Survey: 2006-2010 and FBI reports."

Sounds good.

But wait

One of the key challenges about sexual assault statistics is that it’s nearly impossible to gather accurate and consistent data about incidence and prevalence.  This infographic doesn’t do a perfect job, but it combines data from several sources, both domestic and international.

* * * * *

For those of you who have asked, here is the background on the stats we used:

  • 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)

No links, nothing. We have to trust Beaulieu that this is what the statistics show. 

I know that it's not politically correct to call "bullshit" when victimocrats are talking about rape, but bullshit.

It's no longer "National Crime Victimization Survey: 2006-2010 and FBI reports" but "data from several sources, both domestic and international," which suggests that the previous explanation, "Statistics from Justice Department, National Crime Victimization Survey: 2006-2010 and FBI reports," was untrue.

How many rapes are reported? Maybe 5%, maybe close to half. Maybe more. Beaulieu said that they "assumed" 10%; it's dramatic.

Of those reported cases, how many are prosecuted ("faced trial")? Nine? No way. "We meant 9%"? Yeah, you've already proven that you're statistical morons. The infographic shows 30% of the reported cases being prosecuted, assuming that someone has to be prosecuted to be jailed.

Of those nine prosecutions, how many resulted in felony convictions ("jailed")? Five, based on yet another assumption, and more undisclosed data, and again the infographic (showing that  1/3 of rapists who face trial are jailed) does not match the number provided. (If you assumed, instead, that Harris County is typical, you would find that in 2012, 172 sexual-assault-of-an-adult cases ended in dismissal or acquittal, and 382 ended in conviction or deferred-adjudication probation—2/3 held responsible, near as dammit. Texas Office of Court Administration. See what I did there?) 

Finally, where does the 2% number come from? Explicitly, still another assumption. You can find a bigger and more credible number, 5.9%, here—certainly not a hotbed of rape apologists. (Note that this statistic is based on allegations being proven untrue. An unproven allegation—a "not guilty," for example—would count as true.) If you wanted a credible reason to assume that the number was even bigger—25%—you could find it here: "Forensic DNA typing laboratories — as numerous commentators have noted — encounter rates of exclusion of suspected attackers in close to 25 percent of cases."

The Enliven Project's infographic is nonsense. If there's a number in it that is anywhere close to correct, it's purely coincidental—Beaulieu links to nothing that supports any of their assumptions.

I'm not interested in making up my own nonsense, but if you did not assume that the unproven cases and the not-guilties are true allegations, or if you assumed that the false-positive rate in all allegations matched the exclusion rate in cases with DNA submitted to laboratories, then that infographic would look very different.

You know what I think happened? I think someone sent Beaulieu an email like this:


I wanted to drop you a quick email regarding your site at and ask whether you would be interested in us making an infographic for you?

I’m sure you receive several similar guest post requests each week, so I wanted to quickly point out what I’m proposing and why this would be of value to you:

· We’ll provide you with an exclusive infographic created by us. Rather than simply sending you a text based article, we do all of the research and arrange for one of our designers to create an infographic solely for use on your site. This isn’t something that has been or will be published elsewhere.

· Some examples of infographics that we have recently designed and placed include…

Beaulieu bit, and is now scrambling to justify publishing this nonsense.

I don't blame her. I blame the American educational system, which has created a nation of statistically innumerate consumers of easily digested but false information, including her and everyone who unquestioningly spread this graphic around the internet.

We are all dumber for her efforts.


[A side note: in a sympathetic and only marginally better-sourced critique of the infographic at Slate, Amanda Marcotte writes: 

Those who do report run a very high chance of never seeing a conviction, some because police drop the case on the slut-and-liar grounds and some because juries buy the defense attorney's claim that the victim bizarrely preferred being publicly accused of being a slut and liar to quietly forgetting about a night of forced sex.

Marcotte's premise is that women make false allegations of rape after calm deliberation. I believe the premise to be false.

Based on my experience defending what investigation has convinced me are actual false allegations of rape (probably half the sexual-assault-of-an-adult cases I've defended, though you cannot, of course, assume either that I'm objective or that if I were this number would be in any way representative of the entire population of rape allegations), I do not believe that this is true. Here's why:

The false rape allegation is commonly made right after a controlling, possibly dangerous, boyfriend discovers what appears to him to be infidelity. It's a sick, twisted, impulsive way for a person to get out of trouble, and the person who would make such an allegation doesn't spend a lot of time beforehand considering where the allegation is going to go. I wouldn't call them sluts, but there are a whole lot of liars in the world, and once they've told that particular lie it is hard to take it back.

Since I am not claiming objectivity, I invite you to measure that against your experience of human beings and of the world.]



The Honest Courtesan, The Truth About "The Truth About…"

Overlawyered, An in-faux-graphic on rape statistics

Simple Justice, Forever Wrong (Update — and Wronger x2)

The Legal Satyricon, Updated: That “info” graphic about rape? Bend it over, slap its ass, and forcibly fuck it with facts

Me, A Lie Told Often Enough…


Proudly powered by WordPress
Theme: Esquire by Matthew Buchanan.