The table below reports on the results of original research on the registries in fourteen US jurisdictions, involving direct counts of registry records, and comparing the results to published figures.

The results found that the published figures in 2007, on average, over-reported the number of registrants in the tested jurisdictions by approximately 23,000 out of a total of 157,000, or by 13 percent. If that figure can be generalized to the national total reported in 2008 (see related background report, Published Counts of the US Registries), that total should be reduced from about 640,000 to 557,000. Further research is needed to determine a reasonably accurate count.

In applying the same correction to the derived measures (% and Men per columns in the Published Counts table), the percentage of the US population that is on the registries remains at approximately 0.2 percent (adjusts from 0.21 to 0.19%) and the number of adult men in the population per adult male on the registry (men per registered man) adjusts from 190 to 220. This means that approximately one out of every 220 adult men in the United States is on the registry.

See the Explanation and Sources of the Table below.

Explanation and Sources of the Table

This section explains the data presented in the above table, Count Analysis of the US Registries, and discusses their sources.

For the Registry and Published columns, see the information in the explanation of the table of published counts. The counts in the Published column, as well as their color scheme, are the same as the corresponding data in the 2007 column of that table.

The Comparative Analysis columns present original data collected for this report. Fourteen jurisdictions have been analyzed so far.

The Registrants column presents original data collected on fourteen jurisdictions for this report. For six jurisdictions, a collection of all registry records available was downloaded from the public website and those records were counted directly (AK, AL, AR, AZ, GA, HI, KS). For five jurisdictions, the numbers of registrants were determined from totals given on the respective registry websites (CA, CT, DC) or in communication with the responsible jurisdiction agency (FL), or both (DE). For the remaining three jurisdictions, a combination of publicly stated counts and database counting was used (CO, HI, ID. For Colorado, the number of records on the public website was taken as stated there, while the number of people on the complete registry was obtained by counting the records on a CD procured from the state.). See the Source Notes column in the table for more details specific to each jurisdiction.

The Registrants column has two numbers in each cell. The top one is the number of people Shown on the public registry website for the jurisdiction, while the second is the Total number of people on the jurisdiction’s full registry, which is available to law enforcement and possibly for inspection by the public. In most cases, these people are all designated as “sex offenders” or “sexually violent predators.” However in some jurisdictions, the numbers also include people who committed only non-sexual crimes, such as child kidnapping in Alaska and violent or drug crimes in Kansas. If Registrants: Total is indicated as “unknown,” that means the information on the registry for that jurisdiction says that some registered people are not included in the public listing, but does not give any way to know how many such people there are. If the Total is indicated as “same,” that means the information provided does not say that any registrants are left out of the website listing.

The data presented in the Registrants column were mostly collected in August 2007, which is the reason that they are being compared to registry counts published in 2007.

The Diff column gives the difference between the Published numbers and the Registrants: Total numbers, where applicable. If the Registrants: Total number is known, either as a separate number or as being the same as the Registrants: Shown, then it is subtracted from the average of the NCMEC and PML numbers for 2007. The difference is given as both a raw number and as a percentage of the 2007 average count. If the difference is a positive number, meaning that the published numbers are too large, then the difference is shown in red. If the percentage is 10 percent or greater, then it is shown in bold face.

The Data Notes column contains information that may explain the reasons for the differences. This column provides information from the jurisdiction about who is included and excluded from the registry and from the public website. It also explains how the data in the Registrants column were collected and calculated. Red type in this column indicates notes of problems that cause inaccuracies in the jurisdiction’s count of people on the registry. (See the types of problems listed in the primary report, Counting and Over-Counting the Registry.) The text is in red bold if the problem is due to multiple listing of individual registrants, or accounts for 20 percent of the count or more, or is of unknown magnitude.

The Source Notes column gives information about the procurement of data specific to the jurisdiction.