Sex Offender Recidivism Fact Based Research Statistics
AVNETNEWS – The following Sex Offender Recidivism Fact Based Research Statistics show that “facts are stubborn things”, refuting the shill, rants and other myths, such as the unsupported “frighteningly high recidivism”. We must read this and understand these laws hoping to protect kids are in reality destroying kids and their families,
A Perspective on Recidivism
Most studies of recidivism begin with the statistics gathered on re-offending. Typically, these figures are assembled by states over three year periods, and this data is appropriately criticized for two principle reasons—that these figures do not reflect accurately what happens over a longer period of time, and that they cannot reflect crimes that are not prosecuted. But these studies do provide a tool by which to compare sex offender recidivism to recidivism for other crimes. And the recent sex offender recidivism numbers (which are included further down here) are lower than any other category of crime.
Persons working in the field—treatment providers and parole officers particularly—tend to rely more heavily on actuarial tools developed that provide predictions about the likelihood of reoffending. Not surprisingly, these tools yield higher rates than the three-year recidivism studies, but rates are often quoted in the 20% range
The myth of very high recidivism seems to rest on a handful of poorly designed studies. I include them here.
“Lifetime Sex Offender Recidivism: A 25 year Follow-Up Study”
320 sex offenders referred to a single clinic for eval between 1966 and 1974
|61.1% sex crime recidivism
88.3% including confessions in counseling and new arrests that did not result in conviction; another measure that only included offenders outside their own family yielded a 94.1% rate
|Criticized as a non-random sample limited to a group who were referred for major prosecutions (“the worst of the worst”).
The sample only dealt with period before prisons included a vastly wider range of offenders now arguably less prone to recidivism.
Study eliminated everyone whose records were lost or purged from system after 15 years because of no new crimes or charges, eliminating most non-recidivists from the study
Since at least 50% of the sample were already recidivists by Langevin’s definition, obviously he would dramatically inflate his figures, since the commonly accepted definition of recidivism is a new crime committed after release
In response to his critics, Langevin himself has cautioned against making claims about all sex offenders based on his data.
Study of 136 rapists and 115 child molesters released from Bridgewater civil commitment center in Mass. 1959-1986.
|32% sex crime recidivism for molesters; 25% for rapists (over widely varied periods).
Over 25 years, he estimated rates of 52% and 39%
|Covers approximately the same period as Langevin study
Also only considers narrow range of offenders; average child molesters had 3.6 offenses already and average rapist had 2.5 offenses. By Langevin’s definition, the recidivism rate would have been almost 100% Even the Prentky team said, “The obvious marked homogeneity of sexual offenders [in this sample] precludes automatic generalization of the rates reported here to other samples.”
|2000 and 2009||Andres Hernandes study at Butner FCI – sample of 155 child porn offenders in treatment||85% admitted in treatment to a previous hands-on offense, suggesting that child porn users overwhelmingly are also hands-on abusers||The methods of the study are widely criticized for its poor research model. Confessions were coerced by the threat of dismissal from the program with resultant return to the general population bearing the stamp of sex offender. Like the Langevin study, this study also raises serious questions about definitions of recidivism. The study has been dismissed by judges as lacking credibility on multiple grounds. The 2000 study was never accepted for publication; although the 2009 article was accepted for publication by The Journal of Family Violence, the BOP requested that the article be withdrawn (it was not), and a bureau official wrote: “We believe it unwise to generalize from limited observations gained in treatment or in records review to the broader population of persons who engage in such behavior.” The journal’s peer review process has also been criticized as remarkably lax—it is alleged that the journal allows authors to suggest their peer reviewers and to blackball reviewers they wish to avoid. Thus far, we have not confirmed that charge.|
Comparisons of who actually commits new sex crimes
According to a recent New York study:
- 95.88% of arrests for all registerable sex crimes are of persons previously non-convicted of a sex offense.
- 95.94% of arrests for rape are of previous non sex-offenders.
- 94.12% of arrests for child molestation are of previous non sex-offenders.
(Sandler, Jeffrey C, et. Al., Does a Watched Pot Boil? A Time-Series Analysis of New York State’s Sex Offender Registration and Notification Law. Psychology, Public Policy, and Law, 2008 Vol. 14, No. 4, 290)
These results are closely parallel to figures compiled by the US Department of Justice that show that 93% of child sex abuse is committed by a person whom the child knows. In 47% of the cases, the perpetrator is a member of the family. Only 7% of offenses are committed by strangers.
If sex offender recidivism were not exceptionally low, these figures could not be this high. The hugely disproportionate number of sex offenses committed by previously non-convicted persons raises questions about the utility and justice of a registry which subjects over 20,000 Oregonians to crippling restrictions, none of which address the source of the vast majority of sexual abuse.
Recent Reported Rates of Sex Offender Recidivism*
Most of these studies are based on the standard 3 year reporting system. More details are provided on the next page, including exceptions to the 3 year period. These recent rates are noticeably lower than reported rates a decade ago. This difference probably reflects at least two factors:
- crime rates in general have been falling during this period, and sex
offender recidivism seems to reflect that trend;
2) current studies appear to include a larger and different demographic than a decade or so earlier when both statutes and enforcement procedures were substantially different.
In all cases but two here, the rates are 4% or lower. In more than half, they are 3% or lower. Taken together, these figures seem to underline the reason that the researchers in the last study (Connecticut) wrote:
These low re-offense rates appear to contradict a conventional wisdom that sex offenders have very high sexual re-offense rates. The real challenge for public agencies is to determine the level of risk which specific offenders pose the public. [emphasis added]
* All rates come from studies explained with references on the following pages.
Studies since 2007
|2007||Missouri DOC study of 2,777 offenders released from 1998-2007||1.9% after 3 yrs.; 3.5% after 5 yrs.||These rates were 1.1% and 2.7% for offenders who completed MOSOP, a mandated treatment program.|
|2007||Alaska Judicial Council report||3% in 3 yrs.||Sex offenders had the lowest rate of reconviction for category of offense|
|2007||Minnesota DOC study 3,166 offenders released between 1990 and 2002||3% in 3 yrs. in 2002||10% av. rate over 8.4 yrs; in 1990, rate was
17% in 3 yrs. This might suggest that rates are falling substantially in recent decades.
|2007||Jared Bauer of West Virginia DOC 325 offenders tracked for 3 years from 2001, 2002 and 2003||1% for sex crime w/ victim||This rate might be inaccurate, since victimless crimes are not included, although there is little evidence to suggest that the other cohort would have a higher rate. 9.5% of sample returned to prison for other reasons.|
|2008||California SOMB report||3.55% in 3 years
|Cites figures from CDCR for prisoners released in 2003|
|2009||Endrass et al—Swiss study of 431 users of underage porn||.8% w/ hands-on sex offense; 3.9 % w/ hands-off sex offense||Study covered 6 years; concluded that child pornography alone is not a risk factor for hands-on offenses.|
|2009||Orchowski and Iwama study of offenders released in 2001 for US Justice and Research Association||AK – 3.4%
AZ – 2.3%
DE – 3.8%
SC – 4.0%
IL – 2.4%
UT – 9.0%
|Comparison national 3 yr. rate was 5.3% for offenders released in 1994
|2010||Maine study tracked releases for 2004-2006||3.8% in 3 yrs.|
|2010||California CDCR report||5% in 3 yrs.|
|2012||Connecticut Criminal Justice and Planning Division Study of 746 offenders released in 2005||1.7% returned to prison||Study covered 5 years, not 3. It showed 3.6% arrests; 2.7% convictions.
2007 Missouri Study
2007 Alaska Study
2007 Minnesota Study
2007 Jared Bauer West Virginia study
2008 California SOMB study
2009 Indiana study
2009 Endrass et al study
2009 Orchowski and Iwama study
2010 California CDCR report
2010 Maine study