Saturday, April 20, 2013

Fisking a rape statistic

This site caught my eye.

And here's the claim that really had me curious: Factoring in unreported rapes, about 6% of rapists will ever spend a day in jail.

Interesting stat. Let's see how they run their numbers:

  • 60% of rapes/sexual assaults are not reported to police, according to a statistical average of the past 5 years.  Those rapists, of course, never spend a day in prison.  Factoring in unreported rapes, only about 6% of rapists ever serve a day in jail.
  • If a rape is reported, there is a 50.8% chance of an arrest.
  • If an arrest is made, there is an 80% chance of prosecution.
  • If there is a prosecution, there is a 58% chance of conviction.
  • If there is a felony conviction, there is a 69% chance the convict will spend time in jail.
  • So even in the 39% of attacks that are reported to police, there is only a 16.3% chance the rapist will end up in prison.
  • Factoring in unreported rapes, about 6% of rapists will ever spend a day in jail.
If you run the numbers, you pretty well come out to around 6.5%. But man... something seems wrong here.

First, look at bullet point one: 60% of rapes/sexual assaults are not reported. To understand why it's a problem to casually group 'rape' and 'sexual assault' together, let's take a good look at the big list of sexual assault definitions on this site:

  • Being bathed in a way that feels intrusive,
  • Being ridiculed or teased about your body
  • Being told all you are good for is sex
  • Being involved in child prostitution or pornography
  • Being shown sexually explicit movies
  • Being made to pose for seductive or sexual photographs
  • Being subjected to unnecessary or intrusive medical treatments
  • Being kissed in a lingering or intimate way by an authority figure, adult, or older youth
  • Having your breast, abdomen, genital area, inner thigh, or butt fondled, or being asked or forced to do this by an authority figure, adult, or older youth
  • Being involved in oral genital contact
  • Experiencing finger or object penetration
  • Experiencing penetration of the rectum, and/or vaginal area
  • Being exposed to adult talk about specific sexual acts
  • Being exposed to chronic nudity, or nudity at inappropriate times
  • Being forced or exposed to hearing or watching adults or teens having sex
  • Being exposed to sexual name-calling
  • Living in constant fear of sexual abuse occurring in your life
  • Lack of privacy in bathing or changing rooms
  • Being watched
This runs quite the gamut, doesn't it? 'Being exposed to adult talk about specific sexual acts' and 'Being exposed to sexual name-calling' is grouped in with 'being involved in child prostitution and pornography' or 'being subjected to unnecessary or intrusive medical treatments'. So being called a slut or a cocksucker is categorized here right alongside your doctor fingering you while you're drugged.

Now, that's questionable enough on its own. But if the 60% statistic is being culled by filing rape and sexual assault under the same heading, noticing a 60% lack of report, and then assuming that 40% of rapes are going unreported... that's a shot in the dark. It'd be a little like counting the number of murders and assaults that go unreported, and then using that number to estimate the number of unreported murders. If the FBI/crime statistics are using a definition of 'sexual assault' equivalent to the site's, I'm going to get worried.

But there's something more unsettling going on here. Let's run the numbers.

Starting off with the poorly supported 40% reporting claim for rapes, we then see that 50.8% of these will result in arrest. Okay, that brings us to 20.32. 80% of those will be prosecuted - now we're at 16.256. 58% chance at conviction - okay, 9.429%.  69% chance at jail time gives us 6.51%. (Doing some rounding up here.) Seems solid, right? Math is math.

Here's the problem: not only is the information about unreported rapes, at least by the page's numbers, suspect... but the numbers, in order to be solid, require the unspoken assumption that 100% of all reported rapes are actual rapes, and that 100% of the 60% unreported claims of rape/sexual assaults are also legitimate.

That bears repeating: the numbers I just ran, in order to arrive at the claim that "only 6.5% of rapists will ever serve jail time", requires that 100% of all reported rapes are legitimate, such that if someone is found not guilty, then the state just let a rapist walk free.

I'm not going to comment more about that right now. I just want it to sink in, so anyone who finds this page can understand exactly what is going on with the statistic I just quoted, and why it may be a something to take with a grain of salt.

4 comments:

Syllabus said...

Notice also how, when the 60% figure is being spoken of, the "rapist/sexual assault" terminology is being used. But, when the 6% number has been derived, it's suddenly only 'rapist'. That seems a little bit inconsistent and suspect to me.

Crude said...

I was trying to give them the benefit of the doubt there, but yeah, that's another slippery point.

To me, what's really insidious is this unspoken assumption that if someone is charged with rape/sexual assault, and if they're found not guilty or not prosecuted, etc, well, then a rapist just walked free.

Imagine if the situation was reversed: someone produces a statistic showing how many men are accused of rape versus how many actually go to trial and are found guilty, and for every 'not enough evidence' or 'not guilty' verdict, we say we have one lying slut ferreted out. It's about as just.

Drew said...

I don't think they're logically making the assumption that you say. But their percentages are indeed suspect.

For example, if we assumed that 58% of rape defendants were convicted as they seem to do (even though it is likely false), then it could still very well be that the 42% acquitted were actually innocent. And you could say the same about several of the other stats. What reason in the world is there to think that a large number of guilty rapists manage to fool the courts?

Also, I don't believe that any significant portion of rapes go unreported. Their percentages are off right from the beginning.

Crude said...

I don't think they're logically making the assumption that you say.

You're going to have to tell me another way to give their numbers if you don't think they're making the assumption I take them to be. It falls out of their calculations and their claims - when they say '6% of rapists will ever spend a day in jail', the only way to state that number is to assume that the 94% left out are rapists too.

For example, if we assumed that 58% of rape defendants were convicted as they seem to do (even though it is likely false), then it could still very well be that the 42% acquitted were actually innocent.

That's exactly what I'm pointing out here. Part of the reason I looked into this was being involved in a conversation lately and being flat out told that some guy who was found innocent of rape/sexual abuse 'was probably guilty of something anyway'.

Also, I don't believe that any significant portion of rapes go unreported.

That's suspect as well, and part of the problem is that even if those numbers are being touted, they are uninvestigated claims of rape/sexual abuse. But why take these as being rapes?