The Supreme Court's Crucial Mistake About Sex Crime Statistics | Hacker News Reader

The Supreme Court's Crucial Mistake About Sex Crime Statistics

casetext.com

61 pointsminapurna28 comments

This isn't as rare as I would like it. A few gun studies from the west coast in the 90's pulled this kind of quote without justification or attribution.

Two additional things to be aware of are conditions of the study and people pulling a statistic without the full qualification.

"This activity reduced X by 50%" is a popular one. It gets used when writing grants, but often the original study had very specific environmental conditions to go along with that reduction. Those are often ignored in the activities spread.

Quite a lot of people pull quotes from even legitimate studies, but forget to also pull the qualifiers. This happens a lot with medical studies.

There is another, but I can hardly blame people giving its an outright deception. Sometime government agencies or individuals falsify data. We all remember the whole vaccine thing and the damage it did. I can now blame folks for repeating this long after the facts are know.

A popular one I've seen on HN is people quoting AMTRAK's budget showing AMTRAK's ACELA Express is profitable. Sadly, that's only true if you use some really shady accounting[1] and completely forget capital costs. You wouldn't know that from the various articles quoting the actual budget report.

Politics sometimes trumps the doing of basic fact checking.

1) here is one of the internal critiques of even the newer accounting https://www.oig.dot.gov/sites/default/files/Amtrak's%20New%2... and http://www.amtrak.com/ccurl/782/517/Amtrak-Report-on-Interna...

I've (sadly) conditioned myself to reading post hoc ergo propter hoc statements as effectively useless in today's time.

There are some statements that I suspect make intuitive conclusions, and many of which I suspect are right (e.g., availability of birth control lowers the rate of abortions and unplanned pregnancies), but even those with which I agree are relegated to the portion of my mind that holds the "do not recite this" category of statistics.

I understand why people are so susceptible to it, and indeed, what makes the scientific method so damn hard to being with, is because isolation and controls in the real world are indeed very hard to come by, but I do wish that it was something we could do less.

Unless I am missing some hidden biases this seems an interesting discussion with some worthwhile thinking material. It does not seem like a "rape apologist", whatever that is.

For me the main argument is a big one::

When the California Corrections Department recently examined cases of sex offender registrants returned to prison for a new offense, they found that in 88% of the cases, the new offense was a parole violation. Parole violations are generally acts that aren’t crimes for anyone not on parole—things like going to a bar or visiting a friend who’s also an ex-felon. Only 1.8% of those re-incarcerated had committed a new sex offense.

That's a really big mis step in defining recidivist rates - and I would think it is worth following up in all cases (not just sex offences)

Eh, I'd say the bigger bit is compared to all ex-felons:

> About 3% of felons with no known history of sex offenses commit one within 4.5 years of their release.

Its pretty clear:

To justify having a subset of ex-felons [sex offenders] on the list, all ex-felons have to be on the list. Otherwise, the justification [they are sex offenders] falls flat when other types of ex-felons commit sex offenses more frequently.

Honestly, it sounds like a publicly auditable and widely accessible list of felons is likely what we need if we consider 1.8% reason to put people on a list.

Ok I get your point, and you seem to be arguing the same as the article - the value of a register of sex offenders is not as great as it might seem.

it depends I suppose on whether the register has much of an effect. I mean if you reoffend even after going to jail, having your name on a list seems unlikely to help.

It isn't identical to the article.

I think there needs to be a list of convicted felons somewhere that is widely accessible with their addresses so people are aware they are in the neighborhood. However, I don't think it should list them as "sex offenders" or the like. There is an unearned stigma in many of the cases. [e.g. urinating in public gets you on THE LIST]

I found this quote in particular to be troubling for the _opposite_ reason. That is, in the case of a sex offender who is a pederast, hanging out with a child can be (and for good reason) a parole offense. The author buries this in his argument in order to hide the information from the audience.

I had not actually considered that. I would then look at reforms of the parole system. In the same way there is murder and manslaughter, there is hanging round school gates and drinking in a bar with other felons.

But yes, good point. Inwonder how the parole violations are recorded?

Better be careful posting this. You don't want to be labeled by Matthew Garrett as a rape apologist[0].

[0] http://mjg59.dreamwidth.org/18505.html?thread=693321

It truly is astonishing sometimes just what a remarkably thought-terminating taboo the topic of rape and sexual assault is. Of course it's a gravely serious one, but I don't exaggerate when I say most people are more comfortable talking about genocide than they are about rape.

Is it because a mass atrocity is more impersonal, or is there a degree of puritanism to the whole angle?

That said, mjg59 has always been a vested partisan on these topics, so I wouldn't use him to make any broader points.

Because nobody will suspect you of having both the intention and power to carry out genocide.

Unless you're white and male, and then you'll be accused by the radical fringe as wanting to rape everything you come across and kill everyone not like you.

So because it's more impersonal, then.

I definitely think that's a big part of it. A lot of the people involved in these discussions probably have been sexually assaulted, but relatively few have had their lives affected by genocide (or even murder).

Puritanism probably plays a role, too. Even consensual sex is often considered an inappropriate topic, so it wouldn't be surprising if that applied doubly to rape.

Consider another explanation: If you use the term "rape culture", you essentially claim that in today's society, oftentimes rape will be accepted (FWIW, I agree with this statement). However, there appears to be a consensus in today's society that genocide is always a bad thing; generally that opinion is expressed without any qualifiers (as opposed to qualifiers like "legitimate rape" etc.)

Lets just start with "rape" being used as an all encompassing term for all sexual crimes, which can get very confusing, after all it happens everyday where someone willfully engages in sex and is simultaneously the victim of rape (e.g. statutory rape). So it is a word covering a vast topic.

Historically, victims of rape were more inclined to keep it to themselves than tell someone, because they were not believed; when they were believed nothing happened to the perpetrator; and sometimes the victims were made out to be deserving of the rape. This is not just true of women, but there are many stories of children who told their parents that people at their churches touched them and then the parents did not believe their own children (double victimization).

Lets take a single modern day accusation...Jameis Winston accused of rape by a freshman college girl. Just look online and you will find opinions on everything from: people assuming he is guilty because he is black she is white; people feeling its just another white girl making up lies about a successful black man; that the redneck town of Tallahassee was persecuting a poor black kid; that the good ol' boys club helped cover up a rape for the start football player; that she was just a drunk freshman slut or a gold digger; that he later didn't pay for crab legs from publix so he must be a rapist...it just goes on and on. The reality is there is one fact, no one knows what the hell happened, maybe even the two subjects don't know because of alcohol, but everyone has an opinion as to what happened instead of an honest conversation about prevention.

You generally won't have these types of controversies with genocide victims, because...well the victims are dead, and its hard to fake dead. Then again there are still many holocaust deniers around the world. Further, still there are many people who don't know there are active genocides taking place right now, and I'll tell you this I have talked to people about Darfur who claim that is not a genocide and I have to admit that makes me pretty damn uncomfortable.

I don't think this is an appropriate parallel... apples and oranges as it were.

Admittedly, I only skimmed both articles, but this article appears to be about misinformation regarding recidivism rates affecting the judgment of legislators. The article you linked is in regards to someone being confused about what constitutes rape.

Why should OP be scared that somebody may label them as a rape apologist for linking this? For broaching the subject of what happens to people convicted of sex crimes?

I had the same thought as you, having also skimmed the articles. Having now read them more fully, it seems that patzerhacker's (admittedly, somewhat kneejerk) reaction draws closer parallels than I initially thought.

The approximation is here: http://www.codon.org.uk/~mjg59/ted_mail/0037.html

In that email, Ted Ts'o disputes the common knowledge that 1 in 4 women are raped. It does not, in my opinion, rise to the level of rape apology (though clearly, that is subjective), and attempts to explain that perhaps the "1 in 4 women are raped" statistic (that seems pervasive) is an exaggerated claim. His best piece of evidence, IMO, is that of the women cited in the statistic that results in the 1 in 4 claim, only 1 in 4 of the 1 in 4 even categorize their own reports as "rape". Whether that's right, wrong or indifferent is not mine to say, as I can see merit on both sides of the argument.

The article here, regarding recidivism of sexual offenders is doing the exact same thing as what Ted Ts'o attempts to do, which is to take a popular, but possibly wrong statistic, and give it more context which may render the original statistic moot.

The parallels are indeed close, which makes me glad that I did not initially downvote patzerhacker's comment without reading both articles. That said, I am not savvy to the nuances of either statistic, so I cannot claim to be authoritative on either, but as a non-interested, passive observer, it seems that Ts'o's branding as a rape apologist seems unfounded, from the information that was linked in the article, and Garrett's claims seem overblown, and predicated on the false dichotomy that because Ts'o doesn't believe A, he must believe B, which does not seem to be the case in his email.

I would half-disagree. There are some high level parallels like you mentioned, but the Ted mail is attacking the statistic by it's definitions, whereas this article attacks the statistic directly because it has no basis in reality and simply persists due to the nature of politics. (Publicly trying to correct this misconception would make a judge/lawyer/politician look like a rape apologist, or weak on crime, or sexual predator friendly in headlines)

By my reading of this article, it is also attacking the statistic by its definition.

The implied definition is that recidivism means "repeated the crime for which they were convicted" or "relapsed into criminal behavior". When it clarifies that the highest rate of recidivism is related to merely parole violations, like going to a bar, or visiting a friend who is also an ex-con, it is attempting to clarify the definition of recidivism, or at least our perception of it.

Side note: On this matter, I'm actually somewhat torn. I think that parole can be onerous and unjust, and some parole officers can tenaciously seek out any and all infractions, while on the other hand, parole is the alternative to a longer prison term, and the parolee has agreed to abide by those extra-stringent rules in exchange for an earlier, supervised release.

It makes me sad that people hesitate to discuss verifiable facts for fear that small-minded people might call them nasty names.

I don't think its the fear of getting called nasty names, its the fear of people calling your boss, spouse, etc. and bringing down the modern, low-energy, high-damage version of a mob. When it only take a couple of minutes a day to ruin someone's career or life, the risk is a bit high.

Statistics, both invented and real, are used all the time by people who want a tool to silence those who disagree with them. But two facts make this a dangerous and inaccurate form of argument: one, statistics can only be used to disprove something, and two, different studies often show wildly different results. If you actually dig into the number that people quote, you'll quickly find other studies that report different numbers, use different criteria and show different conclusions. But you'll only hear someone quote a single study, because it's the one that benefits their argument the most.

I think this article conveys a real theme when it comes to using, misusing or listening to someone else totally making sh#t up that only serves their purpose data. Here are hard decisions based on dubious data with real implications to people's lives. I just don't see how this connects to HN, tech or start-ups.

HN isn't limited to either tech or start-ups.

"On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity."

If this article had been about a typical crime, in a typical city, with little else of interest, then perhaps it would be overwhelmingly unsuitable for HN.

I can't speak for anyone else, but the article gratifies my intellectual curiosity. It's fascinating to find out about such an important abuse of information, spanning decades, at the highest levels of the judicial system.

I hope this link doesn't get flagged.

In HN, I've been seeing a lot of links with useful information be unnecessarily flagged by ideologists.

As opposed to those ideologists that post such links?

New account posts article on controversial political subject restricted impact on US, no prior comments, no direct link to hacking and/or start-ups or even technology. Flagged.

Simply because I don't think this is 'hackernews' and for being pre-emptively labeled an 'ideologist'.

I don't closely follow which articles are flagged, but I think that articles with a tech angle or some novel statistical or scientific analysis are favored over general politics. The reason I flag general political articles (eg articles about race or gender issues without any of the above) is that I don't find the discussions interesting, and I think that the posters are coming from a viewpoint of "this stuff is important so it should be discussed everywhere, whether people want to or not"

Reply on news.ycombinator.com