Skip to Main Content

Study Refutes Common Claim of 100,000 Missing Sex Offenders


Additional contacts: 
Andrew J. Harris, Ph.D., University of Massachusetts Lowell, Department of Criminal Justice & Criminology, 978-934-3978,
Jill S. Levenson, Ph.D., Lynn University, Department of Psychology and Human Services, 561-237-7925, 

As Adam Walsh Act Deadline Approaches, Researchers Call for Data-Driven National Sex Offender Policy 

Findings from a newly released study refute the claim that more than 100,000 of the nation’s registered sex offenders are missing and unaccounted for – a figure that has commonly appeared in statements from government officials, Congressional legislation and dozens of media accounts since 2003.   

The release of these findings coincides with the approaching July 27 deadline for states to comply with federal sex offender registration requirements set forth in the 2006 Adam Walsh Child Protection and Safety Act. As of July 18, only eight states have been deemed to have substantially implemented these requirements, and many have called for Congress to revisit key provisions in the law. The “100,000 missing” figure was repeatedly cited in the debates leading to the law’s passage and in its subsequent re-authorization hearings earlier this year. 

The study, published last week by the journal “Criminal Justice Policy Review,” was co-authored by Jill Levenson, a psychology professor at Lynn University in Boca Raton, Fla., and Andrew Harris, a criminal justice professor at the University of Massachusetts Lowell. The new analysis is part of a multi-phase research initiative to provide state and federal lawmakers with data to inform policies concerning the nation’s sex offender registries. In their article, Levenson and Harris trace the origins and evolution of the “100,000 missing” figure and identify multiple fallacies that have surrounded the statistic. 

Drawing upon findings from recent research, Levenson said, “The actual number of non-compliant sex offenders in the United States is most likely between 25,000 and 30,000, most of whom are concentrated in a small number of states.” These include sex offenders designated by states to be transient, homeless, absconded, non-compliant or whose address or whereabouts were otherwise unknown.

Rates of registration noncompliance among registrants living in the community varied greatly, ranging from about 1 percent in some states (e.g. Florida, Minnesota, Iowa, Missouri, North Dakota, Vermont and West Virginia) to about 13 percent (California, Hawaii and Oklahoma).  Nationally, the median rate of noncompliance is about 2.7 percent.

“Differences in how state systems define and capture data make it very difficult to know exactly how many sex offenders have truly gone missing,” said Levenson, “but even using the most inclusive definitions, our data suggest that the 100,000 figure is highly inflated. That figure is three to four times what the number probably is.” 

The statistic, which is often attributed to the National Center for Missing and Exploited Children or the U.S. Marshals Service, actually originated with a survey by a grassroots organization called Parents for Megan’s Law. Analyzing the etiology and political uses of the “100,000 missing” statistic, Levenson and Harris argue that the figure was based on a flawed methodology and a problematic set of assumptions. They further contend that the number’s capacity to evoke strong emotional reactions, coupled with a lack of clarity about what it actually represents, have made it particularly immune to scientific scrutiny and especially prone to misuse and misinterpretation.  

“This is an immensely compelling, headline-grabbing statistic,” said Harris. “Whether it is true or not, it evokes some of our deepest fears, and ends up carrying a lot of currency in the policy arena.” 

Harris added that over-reliance on figures such as the “100,000 missing” have led to approaches that obscure important distinctions among the sex offender population, with potentially serious public safety consequences.  

“There’s no question that there are dangerous individuals out there who need to be closely monitored,” Harris said, “but nobody is served when we exaggerate and over-generalize. We need to be far more precise and strategic in how we think about these problems.” 

While the study concluded that a comprehensive national estimate remains elusive, it draws upon two recent analyses conducted by Harris, Levenson, and their colleagues – the first utilizing a sample of 445,000 sex offender records drawn directly from public registries, and the second involving a nationwide survey of state registry managers. Based on these data, Levenson and Harris suggest that between 25,000 and 30,000 individuals are technically flagged by registration systems as in violation of registration requirements. They caution, however, that it should not be assumed that all of these individuals are fugitives who are unable to be located.  
Levenson and Harris cited administrative and data problems with many state registries and suggest that a portion of recorded non-compliance may be due to record-keeping failures rather than offender behavior.  

“Considering how much attention Congress has paid to reforming the nation’s sex offender registries, one would expect that we’d be basing decisions on solid data,” said Harris. “But the fact is that many important policy decisions are occurring in an information vacuum. We shouldn’t be crafting national policy based on sound-bite statistics.”