New study shows sexual offense recidivism rates lower than previous estimates

By Maynard Law Office; reprinted with permission . . . Recently, the Bureau of Justice Statistics released a report entitled, “Recidivism of Sex Offenders Released from State Prison: A 9-Year Follow Up (2005-2014).”

“Notwithstanding the sensationalist headline (“three times as likely”), the statistics reported are actually quite favorable.

First, it’s important to note that this BOJ report is not based on samples of offender populations, as is the case in most of the academic research in this area. Rather, the report relies on data from the entire population of prisoners released from state facilities in 2005 (30 state, including NJ, were included in the analysis).

This means that when the report says there is a 7.7% sex offense recidivism rate among those with prior sex offense convictions, it is not an estimate of the recidivism rate based on statistical sampling, it’s the actual rate of recidivism for the population (in this case, as measured by arrests for a new sex offense).

Second, the arrest-based sex offense recidivism rate reported (7.7% over 9 years) is lower than the estimated rates obtained by most meta-analytic studies (ranging from 5-15% over 5 years). This means that the data most of us working in sex offense law have been sharing with prosecutors and the courts are overestimating the actual recidivism rate.

Third, there are other details in the study that impact our presentation of recidivism rates. While felons without a sex offense conviction were less likely to be arrested for a subsequent sex offense (2.3% v. 7.7%), there are a lot more of those ex-felons (381,093) than sex offenders (20,195) being released into the community. So in terms of risk to the public, a citizen is much more likely (six times more likely) to be sexually assaulted by an ex-felon without a sex offense than one with a sex offense.

Fourth, the BOJ study also reports that sex offenders who are arrested are less likely to be convicted (50%) than ex-felons with no sex offense history (70%) suggesting a law enforcement bias leading to more unjustified arrests of sex offenders, perhaps due to their higher visibility to law enforcement because of Megan’s Law. The reality is, in criminology research, it is well established for all types of offenses, that an individual convicted of a particular type of offense is more likely to re-offend with the same type of offense than with a different type of offense, which explains why sex offenders are 3 times more likely to commit a future sex offense than an ex-felon without a sex offense history. Nevertheless, the BOJ study reports that only 16% of all arrests for a sex offense by any ex-felon were committed by a sex offender. 84% of all such arrests were committed by an ex-felon with no sex offense history. Context matters, and statistics can easily mislead the public, law enforcement and even the Courts.

In fact, it was the prior 1997 BOJ study tracking offenders released in 1983 that resulted in Justice Kennedy’s assertion:

“When convicted sex offenders reenter society, they are much more likely than any other type of offender to be rearrested for a new rape or sexual assault.” McKune v. Lile, 536 U.S. 24, 33, 122 S. Ct. 2017, 2024, 153 L. Ed. 2d 47 (2002). While a true statement, it ignored the fact that the reported recidivism rate for rape at the time was 7.7% — far from the 80% recidivism rate Justice Kennedy cited to (and ultimately sourced to a Psychology Today article, not a scientific study) in McKune v. Lile.

Finally, the BOJ study also establishes a baseline rate of sex offending (2.3%) among a non-sex offending population of ex-felons. This is very consistent with the research of Hanson, et al. (Reductions in Risk Based on Time Offense-free in the Community…, Psych., Pub. Ply. & Law, 24, 48-63), which found that over time, sex offenders reach a level of offending indistinguishable from the general population of ex-felons (a level he calls “desistance”) and thus cannot be justifiably treated differently from other ex-felons. For example, most lower risk sex offenders will reach that 2.3% recidivism rate five years after release, calling into question why they are treated differently than the 2.3% of non-sex offending ex-felons not subject to Megan’s Law.Properly analyzed, the 2019 BOJ study provides very useful data to counter prevalent assumptions and myths about sex offending.

The study clearly demonstrates that most sex offense recidivism is committed by young offenders (under 25 years old) and within the first 3 years of release. Rates of re-offense decline steadily over time offense-free in the community, and with age. This data would support various types of as-applied and facial challenges to Megan’s Law-type statutes and can be a useful tool in the arsenal of defense attorneys working in this area.

Help us reach more people by Sharing or Liking this post.


Viewing 5 reply threads
  • Author
    • #56346 Reply

      This report is nothing new to those of us who follow these things closely. The potential problem with it that I see coming is that many won’t read past the sensationalist headline. Certainly not legislators.

      I can’t help but wonder if the release of this report is tied to the deferred ruling in Michigan where the judge ordered the legislature to revise its registry. Seems to me the judge is trying to avoid striking it down as unconstitutional. Also seems rather moot, as the new registry (should it be written and enacted) is not likely to be any more constitutional that the current one.

      The entire scheme, from its origins and intents to implementation, is all based on false premises. Altering, revising, or amending it to make it “effective” or “efficient” is strictly cosmetic, like claiming your car will get better gas mileage if you paint it a different color.

      • #56374 Reply
        Ed C

        Dustin is absolutely correct. Neither legislators trying to gain political points, nor prosecutors concerned only with winning will look deeply into the report. Rather they will take pieces out of context. That is exactly how the “frightening and high” misconception was introduced into legal rhetoric. I have yet to completely digest the report, but one point jumped out at me.

        Footnote 1 says, “For this report, “’sex offenders’ refers to released prisoners whose most serious commitment offense was rape or sexual assault.” On page 2 the report elaborates, “[r]eleased prisoners whose most serious commitment offense was a non-violent sex offense, such as prostitution or pornography, are included with public-order offenders.” Also note this study looks not at sexual re-offenses, but at arrests for “any type of crime.”

        In the popular vernacular, the term “sex offender” is applied to anyone on the registry. The BJS study addresses only a small subset of that population, most notably excluding non-violent and pornography offenses. This study strictly applies only to those convicted of rape or sexual assault. According to a 2016 report from the U.S. Probation and Pretrial Services Office, these are most likely to be rearrested for any type of crime over a 3-year period. That study indicates the rearrest rate for those convicted of sexual assault at 23.1% for a major offense arrest. For persons with a pornography offense, the rate is 13.0%. In the case of pornography offenders, 11.6% of the recidivism is due to probation violations, and only 2.6% for any sex arrest.

        That the BJS study addresses only those convicted of rape or sexual assault is an important takeaway. Legislators and prosecutors will attempt to spin the study to indicate it applies to all those on the registry. Each attempt to do this needs to be decisively shot down to avoid another “frightening and high” debacle.


    • #56418 Reply

      OK so the report was done incorrectly and is basically a lie. Great it is good to know the powers that be are willing to break the rules as usual to keep their worker drone program operational.

    • #56458 Reply

      I am curious to know if there is any data from this report that would show any correlation with reoffenses that occurred where the residential and/or work distance restrictions would have made any difference? Anyone with knowledge please respond. Thanks

      • #56555 Reply


        I would argue that the report itself shows there is no correlation, in that recidivism is so low despite the residence, presence, and employment restrictions in place.

        In broader terms, I Google the phrase “sex offender arrested” every week or so. Nearly all (99% +) hits are for status (parole, probation or registry) offenses. The very few that aren’t (think there were 4 or 5 in the year or so I’ve been checking) had absolutely nothing to do with where the accused lived or worked, social media or internet access, Halloween or Christmas decorations, viewing porn (child porn exempted, especially where the charge resulted from an entrapment sting), or any other restriction that registrants are subjected to. Also, the accused’s status as a registrant is never known until after arrest.

        This particular report is deliberately misleading, specifically using big percentages of small percentages to create a false impression, most certainly by design. Hard numbers draw vastly different conclusions.

      • #56664 Reply
        Ed C

        I’ve referenced three articles below. The first is from the Case Western Reserve Law Review. The other two came from a footnote in that article. Since the posting rules don’t allow hyperlinks or email addresses, I suggest you simply google the titles. I did, and all three popped right up. Veritas.

        1) “The Strings in the Books Ain’t Pulled and Persuaded”: How the Use of Improper Statistics and Unverified Data Corrupts the Judicial Processing Sex Offender Cases (2019)

        2) Kristen Zgoba et al., “Megan’s Law: Assessing the Practical and Monetary Efficacy 37” (2008)

        3) A more recent study by Professor Zgoba and two colleagues tracked 547 convicted sex offenders over a fifteen-year period and concluded that SORN laws “do[] not have a demonstrable effect on future offending.” Kristen M. Zgoba, Wesley G. Jennings & Lara M. Salerno, “Megan’s Law 20 Years Later: An Empirical Analysis and Policy Review”, 45 Crim. Just. & Behav. 1028, 1044 (2018);

    • #57889 Reply

      Lol! Believe me, the recidivism rate could come in at ” .0000001%” and proponents of SO laws would still scream and holler, b#@$h and complain, call for more restrictive legislation at the state and federal level. There is no such thing as a bottom of the well for these people. There is a type mass psychosis going on behind this issue but I can’t put my finger on it. But I focus on the legal aspect of this issue, specifically, federal judges know damn well the SOR violates Due Proces (procedure/substantive) and the main one is the Ex Post Facto clause when applied to a person who’s conviction came before the enactment of the law. In my opinion federal judges are “COWARDS” for not striking down SOR laws on these tenants alone.

    • #57975 Reply
      Derek Logue of

      I can understand why no one gets the latest DoJ study right. The new study only used the number of released prisoners that were currently in jail on a sexual offense; they did not count among the numbers a person with a prior offense but was currently incarcerated for any other crime type upon release.

      The rearrest number for EVERYONE with a prior sexual offense was 7.2% rather than 7.7%. Since the reconviction rate was half that, that means 3.6% of the total sample received a new sex crime conviction in the 9 years following release.

      The problem is the stats are buried deep in other places of the report.

    • #59055 Reply
      michael Kuehl

      To my knowledge, not a single woman who was sentenced to jail or prison and a lifetime or at least 20-30 years of sex-offender registration has “re-offended” by having sex with another young men under statutory age. Their recidivism rate is zero. Visit my website, free Abigail Simon,’ and read the article “recidivism rates of sex offender,” in which I explain why this is so.

Viewing 5 reply threads
Reply To: New study shows sexual offense recidivism rates lower than previous estimates
We welcome a lively discussion with all view points provided that they stay on topic - keeping in mind...

  • *You must be 18 or older to comment.
  • *You must check the "I am not a robot" box and follow the recaptcha instructions.
  • *Your submission must be approved by a NARSOL moderator.
  • *Moderating decisions may be subjective.
  • *Comments arguing about political or religious preferences will be deleted.
  • *Excessively long replies will be rejected, without explanation.
  • *Be polite and courteous. This is a public forum.
  • *Do not post in ALL CAPS.
  • *Stay on topic.
  • *Do not post contact information for yourself or another person.
  • *Please enter a name that does not contain links to other websites.

Your information:

<a href="" title="" rel="" target=""> <blockquote cite=""> <code> <pre class=""> <em> <strong> <del datetime="" cite=""> <ins datetime="" cite=""> <ul> <ol start=""> <li> <img src="" border="" alt="" height="" width="">