13 July 2009

Sex Offender Recidivism Rare In California

California sex offender data shows that sex offender recidivism is rare, and that the vast majority of sex offenders who do reoffend do so in the first three years after they are released.

New research in California shows that only a tiny fraction - 3.38 percent - of released sex offenders are convicted of a new sex offense within 10 years of release. The study followed 3,577 prisoners who were released between 1997 and 2007 after serving time for sex offenses.

In an even larger parallel study by California's Sex Offender Management Board, tracking 4,204 paroled sex offenders, only 3.21 percent were convicted of a new sex offense within 5 years of release.

In both studies, almost all of the recidivism came within the first year post-release. Sex offenders were returned to custody for parole violations at a lower rate than other paroled prisoners, despite the fact that they were supervised more intensely. And they were more likely to be rearrested for crimes other than sex offenses.

The findings are consistent with a smaller study two years ago of recidivism by civilly committed Sexually Violent Predators. Of 93 such high-risk offenders released from Atascadero State Hospital without completing treatment, only 4.3 percent reoffended within six years.

"The data call into question the dramatically higher recidivism rates cited by state evaluators at Sexually Violent Predator (SVP) civil commitment trials. Those data are based on Canadian research with an actuarial instrument called the Static 99. The Static 99 recidivism base rates are 18 percent after five years and 21.3 percent after 10 years, many times higher than the California data."

From here.

The total five year recidivism rate for sex offenders in California was 4.69% (including sex and non-sex offenses), and the total technical parole violation rate over five years was 47.05% (compared with 60-70% for all CDRC parolees, many of whom are less intensely supervised).

The ten year study showed a 3.86% total recidivism rate in ten years (using a smaller data pool) and a 49.06% technical parole violation rate over ten years.

The pattern of recidivism in the five year study involving 4,024 released sex offenders in California was as follows: (sex offense, non-sex offense, technical parole violation):

Year 1: 86, 114, 1455
Year 2: 25, 57, 378
Year 3: 14, 24, 123
Year 4 & 5: 10, 2, 22
Five year total: 135, 197, 1978
% in first two years: 82%, 87%, 93%
% in first three years: 93%, 99%, 99%

The pattern of recidivism in the (overlapping sample) ten year study involving 3,577 released sex offenders in California was as follows (sex offense, non-sex offense, technical parole violation)

Year 1: 79, 89, 1316
Year 2: 26, 42, 291
Year 3: 10, 5, 115
Years 4 and 5 combined: 3, 1, 24
Years 6 through 10 combined: 3, 0, 9
Ten year total: 121, 137, 1755
% in first two years: 87%, 96%, 92%
% in first three years: 95%, 99%, 98%

The study casts serious doubt on lifetime sex offender monitoring statutes. Almost all of the risk of recidivism and parole violations is in the first three years after release, with the first two years being by far the most important.


constitutionalfights said...

We have posted many articles regarding recidivism statistics from the US Dept. of Justice and individual State studies. Check us out at Constitutionalfights.org





Anonymous said...


Anonymous said...

Yeah, I believe in the tracking capabilities of the California Government - they did such a wonderful job of it with Phillip Garrido. How many of these offenders hooked up with women who had their own children as well as nieces, nephews, grand kids? How many of those children have friends they can bring over to the house? How many of the women started a family with a sexual predator? You don't have to re- offend outside if you can keep it inside your home. isn't that what we learned from Garrido?

Anonymous said...

First I think a good point was made in general about the facts that upwards of 93% of sex offenses are NOT committed by strangers but buy person known to the victim.

Here is a quote from the journal Criminal Policy Justice Review:

"Criminal Policy Justice Review March, 2009 Sex Offender Housing Availability pg 107

Finally, residence restrictions are based on the premise that children are at risk
from strangers lurking in schoolyards and playgrounds, despite evidence that 93%
of sexually abused children are molested by family members, friends, and acquain-
tances (Bureau of Justice Statistics, 2000). In the present study as well, the major-
ity of offenders had sexually assaulted victims well known to them. Researchers
in Minnesota found that sex offenders who accessed unknown victims in public
places were much more likely to assault adults than children, and none of the cases
involving minor victims was facilitated by close proximity to schools or parks
(Duwe et al., 2008; Minnesota Department of Corrections, 2003).Thus, residence
restrictions are unlikely to prevent sex crimes. The majority of sex offenders are not rearrested for new sex crimes, and those who commit new offenses typically target known victims, not unfamiliar children at schools or playgrounds."

My second comment / question is in regard to the actual tool used by many states ( New York included )to assess risk ... the Static99 (www.static99.org ). Has a study been made of Static99 results using the the same samples used in the California studies to see if the Static99 is a real predictor of outcomes? If not, it should be looked as a test of it's accuracy since so many decisions are at in part made based on this tool.