Question: What is Attleboro doing about sex offenders? Answer: A hearing will be held Oct. 4 concerning banning sex offenders from certain public places.

While the city council should be commended for being proactive, I am not sure that the proposed solution will have any effect on making people safer.

As of Sept. 14, according to the Sex Offender Registry Board website, there were 71 Level 2, and 19 Level 3 registered sex offenders in Attleboro.

Crime policy should be based on "evidence" of what is effective. But numerous laws are based on what people think might be effective, and when measured for effectiveness, the laws are found to produce no effect on public safety.

Concerning this issue, recent research finds that if a person is likely to reoffend, restrictions on where the person can or can't go may displace the problem behavior. My concern is that this ordinance may just displace the problem. If people are concerned about sex offenders, limiting where they can go leads us to a disproportionate concentration of where they are allowed.

Other recent research finds that: "The data presented here do not support the claim that the public is safer from sex offenders due to community notification laws. The data do, however, provide modest support for a key assumption of notification laws: that children receive more protection against victimization when their families know about a high-risk sex offender residing nearby. What is unclear is the quality and relevance of this increased protection."

This is not to suggest we should not have sex offender registries or restrictions on where sex offenders can or can't go. What it suggests is that sex offender registries and banning registered sex offenders from certain places may provide a false sense of security, and so other strategies are necessary.

Civil confinement is one option to protect the public but it is important to know that civil confinement is about four times more expensive than regular incarceration.

It's hard to believe but sex offenders have the lowest rate of recidivism of all the crime categories. This completely flies in the face of conventional wisdom about sex offenders being the most likely group of criminals to reoffend for their initial crime, but these are the facts. It could be argued that sex offender recidivism isn't detected and that is why this number is so low, but that could also be said of other crime categories, too.

In Massachusetts, the rate of recidivism (reincarcerated) for sex offenders and other crime groups is as follows: Property, 52 percent; Person, 50 percent; Other, 45 percent; Drug, 36 percent; and Sex, 19 percent. Excluding technical violations, i.e. not a new crime, sex offender recidivism is about 2 percent. And when we ask what the rate of recidivism is "for a new sex offense," we find that it is less than 1 percent. This is consistent with national data complied by the Department of Justice. The percentages rearrested (but not necessarily reincarcerated or guilty) for the "same category of offense" for which they were most recently in prison was 2.5 percent of released rapists.

The Bureaus of Justice Statistics found that when sex offenders do recidivate with a sex offense, approximately 75 percent victimize an acquaintance. The important point of this is that this proposed ordinance and current sex offender residential restrictions often don't account for this and many other findings.

Just because sex offenders have a low rate of recidivism it doesn't mean that they aren't dangerous. Dangerousness = Seriousness X Probability. If the probability is low, but the seriousness is high, we still have a problem to deal with.

So what can or should be done?

Independent studies of the effectiveness of in-prison treatment programs for sex offenders have shown that evidence-based programs can reduce recidivism by up to 15 percent. Recidivism can be further reduced up to an additional 30 percent with specific after-prison interventions. However, our current policies make no sense; we release many offenders to the public without some form of post-release supervision. Post-release supervision helps decrease recidivism since it involves keeping an eye on the ex-offender, but also with assisting the ex-offender to find a job, obtain treatment and find housing, all of which are important to staying crime free.

Regardless of the program, it is very important to measure the effect the program has on recidivism; just because something is evidence-based or has popular support we can't just assume it works.

The important thing to note is that we can't make assumptions about what works in public safety based on how we think something is or should be - what works and what doesn't is sometimes counterintuitive.

PAUL HEROUX of Attleboro is a former director of research and planning for the Massachusetts Department of Correction and has a master's in criminology from the University of Pennsylvania. He can be reached at