Laura Ahearn is head of Parents for Megan’s Law, based in Suffolk County, New York. Her advocacy group runs a crime victims center and a program that monitors the accuracy of the photos and addresses of county residents whose names are on the state sex offender registry.

The monitoring program is the only one of its kind in the country. In May 2013, the Suffolk County legislature handed PFML responsibility for knocking on the doors of registered offenders twice a year to make sure that they still live where the registry says they do. No other jurisdiction has farmed out its offender monitoring, a job done everywhere else by cops.

At a recent presentation in the Long Island town of Babylon, Ahearn set out to affirm for a mostly friendly crowd that her monitoring program cuts sex crimes.

Since the contract’s launch, PFML has done more than 9,000 home checks to ensure that registered offenders live where they say they do. That’s resulted in more than a hundred arrests for registry violations. In her talk, Ahearn asserted that when the program began, 20 percent of the offenders whose doors they knocked on were out of compliance with their registration requirements, and 35 percent of level-3 offenders hadn’t provided a recent photo as required.

Why does that matter? Because, she said, failure-to-register violations are linked to sex-offense recidivism. One of her slides noted the following:

Research by the Washington State Institute for Public Policy found that sex offenders with a conviction for failure to register had higher rates of recidivism in all categories when compared to sex offenders without a conviction for a failure to register. Failure to Register (FTR) sex offenders were 54% more likely to be convicted of another felony sex offense compared to non-FTR offenders (4.3% v. 2.8%)…

That sounds impressive. Failure to register sounds like a proxy for future crime. Nail those who are out of compliance–as her group has done a hundred times–and you’ve protected kids.

But her description leaves out a critical point. Here’s how the U.S. Department of Justice—no friend of registrants–describes that same study:

In a Washington State study, WSIPP (2006) found higher recidivism for noncomplying sex offenders compared to their registration-compliant counterparts. Noncomplying sex offenders had a felony sex crime conviction recidivism rate of 4.3 percent, while complying sex offenders had a rate of 2.8 percent. It is unknown whether this difference was statistically significant.

That is: The study that PFML cites as the basis for their program, which Suffolk County funds to the tune of $800,000 a year, is based on a single study, whose results aren’t necessarily statistically significant.

It gets worse. The Department of Justice reports on three other more recent studies, in Minnesota (2010), South Carolina (2009), and New Jersey (2012). None of the three found any significant differences in recidivism between registration-compliant and noncompliant sex offenders. Minnesota study co-author Kristen Zgoba notes that failure-to-register offenses—not actual sexual re-offenses—have become the most common recidivism offense for sex offenders released from Minnesota prisons. In that study, a failure-to-register offense predicted only one thing—another failure-to-register offense, not another sex crime or offense of another type.

There is one item out of all of this that could bear more scrutiny—PFML’s motivations. But Upton Sinclair may have figured that out years ago. “It is difficult to get a man to understand something,” he said, “when his salary depends on his not understanding it.”