Is The Supreme Court Relying On False Data On Sex Offender Recidivism?

Supreme CourtAdam Liptak has a fascinating article out this week on the lack of foundation for a statistic that has been heard in the Supreme Court like a virtual manta:  that sex offenders have an 80 percent high rate of recidivism.  Liptak. however, has researched the source of that statistic and found that it is based on the most casual and unreliable reference from a 1988 book.

Liptak noted that the “high rate of recidivism” of sex offenders was referenced last week and that the Court has repeatedly relied on the 80 percent figure.  He then noted

But there is vanishingly little evidence for the Supreme Court’s assertion that convicted sex offenders commit new offenses at very high rates. The story behind the notion, it turns out, starts with a throwaway line in a glossy magazine.

Justice Anthony M. Kennedy indicated the source of the statistic fromMcKune v. Lile. It was a reference to “A Practitioner’s Guide to Treating the Incarcerated Male Sex Offender,” published in 1988 by the Justice Department.  The 231 page book however has lower cited rates of recidivism and “[o]nly one source claimed an 80 percent rate, and the guide itself said that number might be exaggerated.”

The Court may then have simply repeated the “fact” until it is now a virtual touchstone of every argument and opinion supporting sentencing rules.  It could be an example of legal mythology becoming legal fact.

I strongly recommend Liptak’s article, which can be found here.

28 thoughts on “Is The Supreme Court Relying On False Data On Sex Offender Recidivism?”

  1. There is a cure for these offenders. Corporal punishment. This is not like Sargent punishment or Private punishment. The jury decides. Ten years in jail AND cut off the scrotum and weeny and do it on live television for all to see. And, a tatoo on the forehead: No Balls.

    1. I think the tattoo idea is good for repeated recvidists of child molestation.

      1. Gary T – the Elizabethans branded criminals in obvious spots. It was also done in public.

    2. You’d change your mind quick on that when one of your own family members or maybe a close friend gets charged with a “sex offense.” Twenty-five (25%) of sex offenders are teenage boys. I’ll let you take a scissors and do it yourself if you are so eager.

  2. The scandal here is, as is common, the judiciary. The Supreme Court indubitably employs a corps of librarians who could be tasked with preparing bibliographies for the justices and their clerks. A subscription to Sociological Abstracts (or to the Friends of the Library w/ research privileges at one of the capital’s seven research universities) could be entered as a line-item in the Court’s budget. But noooooooo…..

  3. From the article:
    “… it has helped justify laws that effectively banish registered sex offenders from many aspects of everyday life.”

    ******************************
    I assume it’s a dynamic, rather than a static, process and perhaps the 80% figure was once true and is falling. Therefore, the conclusion I draw from the new data is that the tougher you are on the sex offender the less likely he/she is to reoffend. Thus, tough measures that “banish” them from “many aspects of everyday life” must be working. To that extent, the way we treat sex offenders should be extrapolated to other heinous crimes if we want to drive down recidivism.

    1. See Herrnstein and Wilson. Crime rates are sensitive to surety and celerity of punishment. Severity of punishment, not so much.

    2. Taking a look at other countries, their laws, and their statistics, will easily disprove your theory. Reoffense studies on sex offenders from Canada and the EU also show very low recidivism rates, if not lower. These countries do NOT have public sex offender registries, harsh sentences, or other collateral consequences like “banishment” as we do in the US. The US is taking a failing approach to criminal justice.

      1. Ask any PO and they’ll tell you the creeps are wired wrong and hopelessly trapped in a lifestyle that makes rehab near impossible. It’s a mental defect that has dire consequences. I favor chemical castration or institutionalization.

  4. Rather than tightly regulating abortion, it is men’s bodies that must be tightly controlled. Men control the sex act, are physically stronger, and must suffer the consequences of failure to control their bodies. The need for abortions is directly relative to the failure of men to control their bodies. Placing the entire burden of that failure on women is cowardly and depraved.

    1. If it’s not rape, then women have the same responsibility to regulate their bodies.

        1. Per the FBI, there are about 125,000 rapes per year in this country. For a fertile woman, the mean probability an unprotected sex act will result in a pregnancy is shy of 3%, so you’re looking at no more than 3,500 pregnancies per year derived from rapes. There are currently about 700,000 abortions performed in the United States every year. Pregnancies derived from rape thus account for < 0.5% of all abortions.

          It's almost never rape.

  5. Until the court is provided overwhelming statistical evidence that proves the recidivism rate to be different, then what basis are they to make their decision? Lacking that evidence, this may simply be the 4 out of 5 dentists approach where you don’t want to err on the side of that lone dentist.

  6. Like Justice Thomas’ comment in his concurring opinion (Bradshaw v. Stumpf) that “this court has never hinted, let alone held, that the Due Process clause prevents a state from prosecuting defendants based on inconsistent theories.” Yet Pyle v. Kansas held exactly that, though you have to give it a close reading. The reason you have to give it a close reading is that it was so obvious to the 1942 Pyle court that they didn’t feel the point needed a lot of justification. Ugh.

  7. You see recidivism with all kinds of crime. The question re sex offenses is whether recidivism therein is more frequent than it is for burglary, robbery, drug charges, drunk driving, and assault. (Wagers it probably is not bar for a small and readily identifiable subset. If that be the case, sex offender registries and the attendant civic disabilities should properly have a much smaller population).

  8. News flash! Lawyers aren’t sociologists. Indubitably, a large number are like the new chairman of the DNC, who admitted he went to law school because he was bad at math. Lawyers aren’t librarians either. Assembling a comprehensive bibliography is not their deal. They’re deal is finding the factoids which serve their pre-determined objects.

  9. At the risk of sounding very un-PC, there are sex offenders and there are sex offenders. I do not know whether the cited studies distinguish between the drunk college student who grabs a woman’s butt and the predator. In the former case, the offender will normally learn to moderate his alcohol consumption and not re-offend. The latter offender is aroused by, take your pick, violence, underaged boys, underaged girls, kiddie porn, etc. His recidivism rate is probably close to 100% for the simple reason that he (it is usually a he) enjoys deviant behavior at least as much as non-sex offenders enjoy consensual sex with adults. Celibacy for the offender is at least as difficult as it is for practitioners of legal sex acts. Recidivism is to be expected. If the offender likes, for example, underaged girls, a stint in prison will not change his proclivities. The offender is hard wired for deviant behavior and, to date, medical science has no cure.

    1. The John Jay study of accused priests discovered that most such priests had one accusation against them, and most of the rest two. However, there was a set of 149 priests who had a mean of 20 accusations against them and accounted for one-third of the accusations against priests. People who chastise the bishops don’t realize that in evaluating a given priest, they were typically assessing a single accusation which landed on their desk more than a decade after the supposed incident. (One of the most promiscuous priests was John Geoghan, who was eventually accused by 130 people of having molested them; about 6 accusations had been reported to the chancery at the time he was removed from ministry in 1992).

  10. On balance though I do recognize that certain sex offenders are incorrigible, but unfortunately the sex offense registration policy is to broad and tends to involve those who are likely to only be single incident offenders, such as those who were juveniles at the time of the offense and the victim had consented but was in some cases only a year or two younger than their statutory offender.

  11. A more accurate method for arriving at actual recidivism rates today is more readily available than it was during the 1980s as database management and queries are orders of magnitude broader and expedient. Rather than rely on anecdotal studies to achieve a result. I propose that formatting the data query would be the most involved part and even this would be minor compared to what was done thirty years ago.

    Unfortunately results of studies become at times a form of habit where the premise becomes increasingly credible and questioning the data lessens. It is especially aggravated when it becomes a device to achieve political goals. Sex offenses where the big political fear device twenty years ago and this, coupled with faulty data, led to what I view as draconian laws that stretched the limits of constitutional protections and broke through in many ways. It also fostered the idea of civil commitment after criminal imprisonment, many leading to an almost permanent state of incarceration far in excess for which comparable determinate sentences could be levied.

    I believe this civil commitment had lead to a trend in legislation toward using the civil law when politicians were frustrated by the courts in their overreach of criminalization and sanctioning. In this example civil sanctions are used to circumvent limits to criminal penalties. Examples outside the sex offense realm include sanctioning driver licenses for matters completely removed from suitability to license, such as for lack of child support payments, or defaulting on payments due to the state.

  12. I think it is part of the culture now that recidivism is high for sex offenders. I do not see the SC making a change in that. Every TV show makes its mark on the culture. I think (assuming the figures are wrong) that it will take two or three generations to change minds.

  13. In studying and researching related area police misconduct vs misconduct of society as a whole i fouind the amount of reliable statistics to be sparse. I started looking at the results of the 1993 Act passed under President Clinton which required all LEA at all levels to report almost daily on all things. As a result and without funding the requirement was largely ignored

    CATO Institute started to fill in some blanks again looking at Police Misconduct and hand researched a goodly number of reports over a number of years and a wide ranging geographical area.

    They made attempts to verify all positions claimed and fact verification became a major time consumer. Along witt that they started receive of find general population misconduct information but quickly determined the 1993 law had been shunted aside first due to fundings and then due to a fast very fast emerging federal police presence.

    About the time they stopped their research a new emphasis on police reporting occured leading to JTTF and the use of computers and a lot of dollars spent. I don’t know the outcome of that effort.

    Prior to the National Chiefs of police did a study on high speed pursuit driving which tied the approximate one death per day nationwide with that one activity and their own study came up with two initial conclusions or at the least start points for a serioud study.

    Dividing misconduct or crime into all it’s categories they compared police misconduct to police population. and general misconduct with general population. the percentages remained very even for both in the spectrum of crimes and in the percentage of who committed the crime.

    1 to 1. 5% of the general population and the same 1 to 1.5% of LEA across the list of crime types.

    Which if not indicated pointed at what were the root causes and were they the dame. Education, psychological testing, poor leadership or none, very bad working hours, economic conditions? The idea was finding out what were the common failings and could they be tested upon application for emplymet im the case of police cadets ad rookies

    One other point stood out. Sex Crimes ran a bit over 1.5% compared to nearer 1% for the other crimes.

    Which made me think of those studies and the new police reporting being used in the terrorism effort.

    So I draw no conclusions. except the CATO study did indicate a worthwhile starting and the nationalized information gathering for terrorism purposes might be something worth writing as progam and punching the go button on the brain box…

    Recedivism brings in another important factor ….in light of what you presented.

    http://www.policemisconduct.net/arguing-the-case-for-police-accountability-part-1/

    Arguing The Case For Police Accountability – Part 1

    May 9, 2010 @ 4:56 PM by David Packman

  14. We should have access to sufficient sex offender records to update established recidivism rates. What is the actual rate? Greater than or less than 50%?

    I think few crimes destroy the public’s trust as much as sex offender, with pedophiles the most reviled. “Trust me, I’ve served my time and paid my debt”, convinced exactly zero people, ever, that a convicted pedophile will ever be safe in a community. Even if the recidivism rate is 10%, each crime represents a child’s innocence ruined.

    I think that the risk of recidivism depends on the motive for the crime. Was it compulsion, innate attraction, a psychopathic lack of empathy? There are so many reasons behind each crime, and I suppose the degree of rationality and tension with compulsion determines risk.

Comments are closed.