The table below lists official and unofficial counts of the numbers of people on all the US state registries from 1998 to 2008, indicating discrepancies in the numbers, rates of change, and the “population frequency” of adult male sex offenders.

See the Explanation and Sources of the Table below.

This section explains the data presented in the table, Published Counts of the US Registries and discusses their sources and discrepancies..

The Registry column provides link to the official sex offender registry websites of the United States and its fifty states and the District of Columbia, along with the title of each site.

The Dru Sjodin National Sex Offender Public Website for the United States serves as a portal to information provided by the individual jurisdictions. It provides individual conditions of use for each jurisdiction, which may include links to various parts of the their websites, but it does not provide links to the sites of the individual registries themselves, which would facilitate searching their databases directly. Those registries often provide more information than the national site.

Most of the links to the registries were obtained from the FBI’s National/State Sex Offender Registry list. However, many of the links on that page were not correct as of September 2008 and have been replaced with links that work as of that time. The titles were obtained directly from each site.

The first “+” column indicates registries that are not just for sex offenders, but that also include people convicted of certain violent, nonsexual crimes. A check mark is given for a jurisdiction for which this is the case and the row for the national registry indicates how many jurisdictions do this. The primary source of this information is a notation (pink box) on News & Noteworthy: Articles Concerning Sex Offender Issues by Arthur Parrish, 2006. To the five listed there, Alaska has been added based on intitial analysis, but more work on this issue is needed.

The dated columns provide information from various sources at various times. The first two columns provide the most recent official government figures. Since the federal government has stopped issuing this data, the task of compiling the numbers has been taken up by two citizen’s groups, Parents for Megan’s Law (PML) and the National Center for Missing and Exploited Children (NCMEC). See more information on them below.

The two DoJ, BJS columns give official counts published by the US Department of Justice, Bureau of Justice Statistics, in Summary of State Sex Offender Registries: Automation and Operations, 1998 and Summary of State Sex Offender Registries, 2001. The number for Connecticut was zero in 1998 because registries were maintained locally and there was no information available at the state level. The figure for Massachusetts in 2001 is an estimate that was indicated in a footnote in the DoJ document. Here it is included in the table and in the national total.

There has been no official update of the government data since then. As of September 2008, official information about the registries from the BJS consists of a brief entry in Criminal Record Systems Statistics that says the individual states and other jurisdictions maintain their own registries and submit data to the national registry. No statistics are provided.

The NCMEC / PML columns give counts stated for 2005 through 2007 by the National Center for Missing and Exploited Children and Parents for Megan’s Law. The top number in each cell is from NCMEC; the bottom one from PML. Many of these pairs of numbers are very different, which is indicated by the appearance of the numbers. Red text means that the two numbers differ by 10 percent or more; bold face is used if the difference is at least 20 percent. Black bold face indicates that one number is more than double the other. NCMEC reports that its numbers are obtained by a telephone survey of the registry offices in each jurisdiction. PML does not state the method used to obtain its numbers.

The 2008 columns provide information based on 2008 figures from NCMEC. (As of September 2008, PML had not published numbers for 2008 yet.)

The Per 100k column gives a number calculated by dividing the 2008 registry count by the population of the jurisdiction. This number is provided in the NCMEC data, but is also calculated here directly from census data (2007 Statistical Abstract, US Census Bureau).

The Men per column gives the number of men in the general population for every adult male on the registry, i.e., the number of men per SO men. The national number of 190 means that approximately one out of every 190 men in the United States is a registered sex offender, according to the NCMEC data. This is technically called the “population frequency” among adult males. It is obtained by dividing the adult male population by an estimate of the adult male portion of the registry count and rounding to two significant digits.

Adult male populations were obtained from census data (State Single Year of Age and Sex Population Estimates: April 1, 2000 to July 1, 2006 - RESIDENT, obtained from State population datasets, US Census Bureau, giving 2006 estimates for each jurisdiction broken down by sex and age).

The estimated adult male portion of the registry counts were calculated by multiplying the counts by 90 percent. This percentage estimate was decided as follows. Of seven jurisdictions for which the database of all publicly disclosed offenders was obtained in the comparative analysis, two included date of birth and gender in all records obtained. These were Georgia, with 14,307 offenders, and Kansas, with 3,821. Among these offenders, the age/gender breakdown was: 18 and over, 17,480 males and 512 females, under 18, 133 males and 3 females. The proportion of males 18 and over was thus 96 percent for the publicly disclosed sex offenders in those two jurisdictions. Because other jurisdictions may have different age and gender distributions, and because there may similarly be a different distribution among the records not disclosed to the public, a more conservative estimate of 90 percent was used.

Frequencies less than 200 are shown in red. Those that are below 100 are in bold red. These indicate jurisdictions in which there are less than 100 men for every adult male sex offender, according to NCMEC’s data.

The second “+” column indicates registries that include people who have been deported, incarcerated or have moved to another jurisdiction. A check mark is given for a jurisdiction for which this is the case and the row for the national registry indicates how many jurisdictions do this. The source of this information is the NCMEC map for 2008.

The Changes column indicates the percentage change in populations within each jurisdiction from each count to the next. For 2005 through 2007, the average of the NCMEC and PFM numbers are used in the calculation. Declines are given in red. Changes greater than 10 percent are in bold face.

Published counts of the US registries are summarized and linked in the following table.

Sources of Published Counts
Year	US DoJ	NCMEC	PML
1998	Summary of State Sex Offender Registries: Automation and Operation 1998
2001	Summary of State Sex Offender Registries, 2001
2005		RSO map, May 2005	Megan's Law Report Card
2006		RSO map, March 8, 2006 (2006 column of table on that page. Data not included in table above.) RSO map, October 17, 2006
2007		RSO map, May 16, 2007 (Data not included in table above.) RSO map, November 15, 2007
2008		RSO map, July 17, 2008

NCMEC. The National Center for Missing and Exploited Children publishes once or twice a year a map showing the registered sex offender populations for the 50 states, DC, and various US territories and possessions. The latest map is avilable on their website (linked from their Sex Offenders page), but archived copies of prior maps are linked in the above Sources table.

The NCMEC data are obtained by a phone survey of the registry offices of the individual jurisdictions.

Since 2007, the NCMEC has included counts for US territories and possessions on its maps. These data are not included in the national totals in the table on this page in order to keep those totals consistent with data from other sources. The totals shown on the NCMEC maps since 2007 do include those additional amounts.

The NCMEC is something of a quasi-government agency because it was established by an act of Congress and is granted official duties, privileges, and funding alongside official government agencies by various legislation (such as Adam Walsh Act § 130, 143(b)(4) and (5), 151(a)(1), 629, 639(c)(2)(E)), while being managed as a private organization outside of government control.

PML. Parents for Megan’s Law publishes the Megan’s Law Report Card, which gives the number of registered sex offenders for the 50 states and the District of Columbia for 2005 through 2007. It does not, however, give any information about how or where the numbers were obtained or calculated. An e-mail requesting source information has not been answered.

There are two pairs of adjacent numbers in PML’s data for 2007 that appear to be accidentally switched with each other and those errors have been corrected in the table. To wit, the number given for Delaware was 696, which has been switched with the adjacent number given for DC, 3307; New Hampshire’s number of 11,566 has been switched with the adjacent number for New Jersey, 3,706.