U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics Special Report Prison Rape Elimination Act of 2003 Sexual Violence Reported by Correctional Authorities, 2006 August 2007, NCJ 218914 ------------------------------------------------------ This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.csv) and the full report including tables and graphics in .pdf format are available from: http://www.ojp.usdoj.gov/bjs/abstract/svrca06.htm This report is one in a series. More recent editions may be available. To view a list of all in the series go to http://www.ojp.usdoj.gov/bjs/pubalp2.htm#svrca ------------------------------------------------------ By Allen J. Beck, Ph.D. Paige M. Harrison, BJS Statisticians and Devon B. Adams, Policy Analyst The Prison Rape Elimination Act of 2003 (P.L. 108-79)requires the Bureau of Justice Statistics (BJS) to develop new national data collections on the incidence and prevalence of sexual violence within correctional facilities. This report fulfills the requirement under Sec. 4 (c)(1) of the Act for submission of an annual report on the activities of BJS with respect to prison rape. Between January 1 and June 30, 2007, BJS completed the third annual national survey of administrative records in adult correctional facilities, covering calendar year 2006. Although the results were limited to incidents reported to correctional officials, the survey provides an understanding of what officials know, based on the number of reported allegations, and the outcomes of follow-up investigations.***Footnote 1: A survey of State-operated juvenile systems and privately or locally operated juvenile facilities was also conducted for 2006. Results from the 2005 and 2006 juvenile surveys will be published in a separate report.*** By comparing results of the 2006 survey with those from 2004 and 2005, BJS is able to assess trends in sexual violence for the first time since the Act was passed. Administrative records surveys are one part of BJS's multi- measure, multi-mode implementation strategy. During 2006, BJS completed development and testing of survey methodologies that rely on reports of victimization provided directly by prison and jail inmates, former inmates, and youth held in juvenile facilities. These methodologies rely on self-administered surveys that offer anonymity to victims of sexual violence to report their experiences. Using Audio Computer-Assisted Self- Interview procedures, respondents interact with a computer- administered questionnaire using a touch-screen and follow audio instructions delivered via headphones. (See box on page 2 for a status report and national implementation schedule.) The 2006 administrative records survey provides the basis for the annual statistical review required under the Act. The survey included all Federal and State prison systems and facilities operated by the U.S. military and Immigration and Customs Enforcement. The survey also included representative samples of jail jurisdictions, privately operated adult prisons and jails, and jails in Indian country. Altogether, the administrative survey included facilities housing more than 1.8 million inmates, or 81% of all inmates held in adult facilities in 2006. (See Methodology for detailed sampling description.) As with previous administrative records surveys, the 2006 survey results should not be used to rank systems or facilities. Given the absence of uniform reporting, caution is necessary for accurate interpretation of the survey results. Higher or lower counts among facilities may reflect variations in definitions, reporting capacities, and procedures for recording allegations as opposed to differences in the underlying incidence of sexual violence. In 2004 BJS developed uniform definitions of sexual violence. Incidents of inmate-on-inmate sexual violence were classified as either nonconsensual sexual acts (the most serious violent forms of sexual assault) or abusive sexual contacts (less serious, but unwanted). Incidents involving staff were separated into staff sexual misconduct (any act of a sexual nature directed toward an inmate) or staff sexual harassment (repeated verbal statements of a sexual nature to an inmate). For this report, all such incidents are considered sexual violence. (See Methodology for detailed definitions.) Detailed tabulations of the survey results by system and sampled facility are presented in Appendix tables 1a--4b, available on the BJS web site at . Allegations of sexual violence rose during 2006 Reports of sexual violence varied across systems and sampled facilities, with every State prison system except Alaska and New Mexico reporting at least one allegation of sexual violence. Among the 344 sampled local jail jurisdictions participating in the survey, 161 (47%) reported an allegation. About 52% of the 46 sampled privately operated prisons and jails reported at least one allegation. The 2006 survey recorded 5,605 allegations of sexual violence. Taking into account weights for sampled facilities, the estimated total number of allegations for the Nation was 6,528. Since the Prison Rape Elimination Act was passed in 2003, the estimated number of allegations nationwide has risen by 21% (5,386 in 2004; 6,241 in 2005). Some of the increase may have resulted from adoption of BJS definitions and improved reporting by correctional authorities. -------------------------------------- Collection of victim self-report of sexual violence in prisons and jails underway BJS is working toward full implementation of the Prison Rape Elimination Act. As of June 30, 2007, BJS and its data collection agents had completed all phases of development and testing. BJS has entered into cooperative agreements to collect reports of sexual violence directly from inmates in prisons and jails, former State prison inmates, and youth in State juvenile facilities. BJS has worked extensively with –- Research Triangle International(RTI)(Raleigh, NC) to collect data from inmates in prisons and jails; Westat, Inc. (Rockville, MD) to collect data from adjuducated youth in State and local juvenile facilities; National Opinion Research Center(NORC)(Chicago, IL) to collect data from State inmates under active parole supervision. Though underlying survey methodology and logistical procedures differ with each of these data collections, the measurement strategies are consistent. The surveys consist of an Audio Computer-Assisted Self-Interview (ACASI) in which respondents interact with a computer- administered questionnaire using a touch-screen and follow audio instructions delivered via headphones. The use of ACASI is expected to overcome many limitations of previous research. The following work has been completed or is underway: Prison and jail inmates * Implementation of the National Inmate Survey(NIS)began in April 2007 in 148 prisons and 302 jails. Data collection will continue through December 2007, with about 90,000 completed interviews anticipated. * On August 30, 2007, BJS will consult with corrections administrators and experts in statistical scaling to discuss various measures of sexual violence by which to rank facilities. * Rankings of prison facilities, as required under the Act, are expected in October 2007; jail rankings are expected shortly after the completion of the data collection. Youth in custody * Testing of the National Survey of Youth in Custody (NSYC) was completed in June 2007. The test involved 12 juvenile facilities in 6 States with more than 750 completed interviews. * Results of the testing and plans for implementation will be presented to juvenile administrators, researchers, and other stakeholders at a national workshop on August 28, 2007. * National implementation is expected in early 2008, pending approval from the Office of Management and Budget (OMB). When fully implemented, the NSYC will include about 15,000 adjudicated youth in a sample of 208 State-operated facilities and 48 large non-State facilities (that had an average daily population of 90 or more youth during 2005). Former State prisoners * The Former Prisoner Survey (FPS) was tested in 16 parole offices with 788 former inmates on active parole supervision. In May 2007, the collection was submitted to OMB for review. * The survey will provide a national estimate of the incidence and prevalence of sexual vicitimization based on reports of former State prison inmates. Data will be collected on the totality of the prior term of incarceration, including any time in a police lockup, local jail, State prison, or community correctional facility prior to final discharge. * National implementation will begin in late 2007, pending OMB approval. When fully implemented the survey will include about 16,500 former inmates in a sample of 285 parole offices. -------------------------------------- Expressed as rates, there were 2.91 allegations of sexual violence per 1,000 inmates held in prison, jail, and other adult correctional facilities in 2006, up from 2.46 per 1,000 inmates in 2004. Overall, the rate in State prisons (3.75 per 1,000) was higher than the rate in local jails (2.05 per 1,000). About 36% of the reported allegations of sexual violence in 2006 involved staff sexual misconduct; 34%, inmate-on-inmate nonconsensual sexual acts; 17%, staff sexual harassment; and 13%, inmate-on-inmate abusive sexual contacts. These percentages were nearly unchanged from those reported in 2005. Correctional authorities reported 3,489 allegations of staff sexual misconduct and harassment during 2006, compared to 3,470 during 2005.***Footnote 2: See Sexual Violence Reported by Correctional Authorities, 2005 at .*** Upon investigation, most allegations were unsubstantiated or unfounded The most common outcome of investigations was a determination that the evidence was insufficient to show whether the alleged incident occurred. In 2006 more than half of all allegations (55%) were unsubstantiated; more than a quarter (29%) were unfounded (determined not to have occurred). About a sixth of all allegations (17%)were substantiated. Previous surveys recorded similar outcomes. Based on completed investigations, allegations of staff sexual harassment and inmate-on-inmate nonconsensual sexual acts were less likely to have been substantiated than other types of allegations. During 2006, 7% of allegations of staff sexual harassment and 14% of inmate-on-inmate nonconsensual sexual acts were substantiated, compared to 19% of the allegations of inmate-on-inmate abusive sexual contacts and 25% of the allegations of staff sexual misconduct. Overall, 967 incidents of sexual violence were substantiated in 2006, compared to 885 in 2005. Relative to the number of inmates, the rate of substantiated incidents of sexual violence in 2006 was 4.3 per 10,000 inmates, nearly unchanged from the 4.0 per 10,000 inmates recorded in 2005. Rates were lowest in Federal prisons and privately operated prisons (fewer than 1 in 10,000). Rates of substantiated incidents in State prisons, local jail jurisdictions, and privately operated jails were 4 to 5 times higher. Substantiated incidents were too few to provide reliable estimates for other types of facilities. Surveys reveal consistent patterns of sexual violence in correctional facilities In 2005 and 2006, correctional authorities were asked to provide detailed information on all substantiated incidents of sexual violence on a separate incident form. Authorities reported information on the circumstances of each incident, characteristics of victims and perpetrators, type of pressure or physical force, sanctions imposed, and victim assistance. The two surveys provide a profile of victims and perpetrators and reveal consistent patterns among the substantiated incidents. Data provided on incidents of inmate-on-inmate sexual violence revealed that -- * More than one inmate was reported to have been victimized in 8% of the incidents in 2006 and 4% of those in 2005. * More than one perpetrator was involved in 10% of the incidents in 2006 and 7% of those in 2005. * Males constituted 82% of the victims and 85% of the perpetrators in 2006, compared to 88% of the victims and 91% of the perpetrators in 2005. * In both years, victims were on average younger than perpetrators. In 2006, 44% of victims were age 24 or younger, while 81% of perpetrators were age 25 or older. * In 2006 whites made up 72% of the victims; blacks, 16%; and Hispanics, 9%. Among perpetrators, 39% were white; 49% black; and 10% Hispanics. Similar distributions were observed in 2005. * In both years at least half of inmate-on-inmate sexual violence was interracial: 6% of incidents in 2006 involved a white perpetrator and a non-white victim; 35%, a black perpetrator and a non-black victim; and 8%, a Hispanic perpetrator and a non-Hispanic victim (not shown in table). Most incidents of sexual violence among inmates involve force or threat of force and occur in the victim's cell, in the evening Correctional authorities reported that physical force or threat of force was used in more than half of all substantiated incidents of inmate-on-inmate sexual violence. In 21% of the incidents in 2006, no force was used or threatened. In 30% of incidents, victims were talked into it. In 7% of incidents, victims were offered protection, bribed, or blackmailed. Force or the threat of force was more common among incidents of nonconsensual sexual acts (67%) than among incidents of abusive sexual contact (44%). The victim was held down or restrained in 38% of the incidents of nonconsensual sexual acts during 2006. In 18% of these incidents, the victim was physically harmed or injured. Although abusive sexual contacts are typically less serious forms of assault, the victim was held down or restrained in 28% of the incidents, and the victim was physically harmed or injured in 10% of the incidents (not shown in table). Overall, the use or threat of force was more prevalent among incidents reported in 2006(58%)than in 2005(51%). In addition, the proportion of victims who were injured rose, from 15% in 2005 to 20% in 2006. Anal or rectal tearing was reported at nearly the same levels (5% in 2006 and 6% in 2005). Incidents of sexual violence among inmates occurred most often in a victim's cell or dormitory (totaling more than 70% in both years). By contrast, 17% of the incidents in 2006 and 21% of the incidents in 2005 occurred in a common area, such as a shower or dayroom. In 6% of the incidents in 2006 and 9% in 2005, the location was a program service area, such as a storage room, hallway, laundry, cafeteria, kitchen, or workshop. Sexual violence among inmates was more common in the evening (between 6 p.m. and midnight)than at any other time of the day, exceeding 40% of all incidents in both surveys. In 2005 it was least common between midnight and 6 a.m. (18%); in 2006 it was least common between 6 a.m. and noon (20%). Incidents of sexual violence among inmates, when reported, are typically reported by a victim or another inmate and not by a correctional officer or other staff. In 2006, 83% of the substantiated incidents of inmate sexual violence were reported by the victim or another inmate; in 2005, 91%. Most victims experienced a change in their housing; most inmate perpetrators received solitary confinement The most common response to a reported incident of sexual violence among inmates was to place the victim in administrative segregation or protective custody and to move the perpetrator to solitary confinement or other higher level of custody. Among victims of inmate sexual violence reported in the 2006 survey, 40% were placed in administrative segregation or protective custody; 13% were placed in a medical unit; and 16% were transferred to another facility (not shown in table). Nearly a quarter (24%)experienced no change in their housing or custody level. Among victims in the 2005 survey, nearly the same actions had been taken; a slightly higher percent (32%) had experienced no change in housing. Inmate perpetrators were moved to solitary confinement in 78% of the incidents of sexual violence in 2006, compared to 71% in 2005. Perpetrators of the most serious incidents, nonconsensual sexual acts, were the most likely to receive solitary confinement (81% in 2006, 72% in 2005). Legal sanctions, including arrest, referral for prosecution, or a new sentence, were imposed on perpetrators in 41% of the incidents in 2006 and 51% of the incidents in 2005. This drop was attributed to a decline in legal sanctions imposed on perpetrators of abusive sexual contacts (39% in 2005; 18% in 2006). In addition, perpetrators in 2006 received a combination of other sanctions, including confinement to their cell or room (16%), loss of privileges (20%), placement in a higher custody level (22%), and transfer to another facility (22%). These types of sanctions were imposed in 2005 at similar rates, except for confinement to their cell or room (28%). The sexual relationship "appeared to be willing" in 57% of incidents of staff sexual misconduct and harassment Correctional authorities also reported detailed data on 544 substantiated incidents of staff sexual misconduct and harassment (up from 344 in 2005). To address concerns about the reporting and interpretation of data in the 2005 survey, BJS changed the item related to the nature of the incidents in 2006. The option "Romantic" was replaced by "Sexual relationship between inmate and staff appeared to be willing." The options "Other" and "Level of coercion unknown" were added. In addition, the options were re-ordered from most to least coercive. (See Methodology for further details.) Though these changes were introduced in 2006, the findings remained similar to those reported in 2005. * The sexual relationship "appeared to be willing" in 57% of incidents in 2006. In comparison, the relationship was classified as "romantic" in 68% of the incidents in 2005. * Physical force, abuse of power, or pressure was involved in 7% of the incidents in 2006, compared to 15% of the incidents in 2005. * A third of the incidents in 2006 involved other forms of assault, including sexual harassment (15%), indecent exposure/invasion of privacy(9%), and unwanted touching for sexual gratification (5%). * In 12% of the incidents in 2006, correctional authorities reported the "level of coercion unknown." Other data reported on substantiated incidents of staff sexual misconduct and harassment during 2006 revealed that-- * In more than half of the incidents, the victim(29%)or another inmate (27%) reported the misconduct. * Most incidents occurred outside the inmate's living area: 46% in a program area, 8% in a common area, and 8% outside of the facility. * Incidents occurred most often between noon and 6 p.m. in prisons (53%) and between 6 p.m. and midnight in local jails (74%). * More than one staff member was involved in the sexual misconduct in 2% of the incidents. * Staff had victimized more than one inmate in 18% of the incidents. Female staff more frequently implicated in sexual misconduct in prisons; male staff in local jails In the 2006 survey, characteristics of victims and perpetrators of staff sexual misconduct and harassment differed by type of facility: * In State and Federal prisons, 65% of inmate victims of staff sexual misconduct and harassment were male, while 58% of staff perpetrators were female. * In local jails, 80% of victims were female, while 79% of perpetrators were male. * 49% of staff perpetrators in prisons were age 40 or older, while 65% of victims were under age 35. * 56% of staff perpetrators in jails were age 40 or older, while 86% of victims were under age 35. * Among staff perpetrators in prisons and jails, 71% were white; 20%, black; and 7%, Hispanic. * Among inmate victims, 66% were white; 23%, black; and 8%, Hispanic. * A correctional officer was identified as the perpetrator in 54% of incidents in prisons, and in 98% of incidents in jails. * A contract employee was involved in 17% of the incidents in prisons and in 2% of those in jails. Three-quarters of staff perpetrators in 2006 lost their jobs; 56% were arrested or referred for prosecution Correctional authorities indicated that staff had been discharged or resigned in 77% of substantiated incidents in 2006, compared to 82% in 2005. Staff had also been arrested or referred for prosecution in 56% of incidents (compared to 45% of incidents in 2005). Approximately 10% of staff perpetrators in 2006 were disciplined, transferred or demoted, compared to 17% in 2005. Among the multiple types of sanctions imposed on staff perpetrators during 2006, discharge or resignation was the most common, constituting 79% of incidents in prisons and 74% of those in jails. Jail staff were more likely than prison staff to be arrested or prosecuted (73% versus 43%). Half of inmates involved in staff sexual misconduct were transferred or placed in segregation Correctional authorities indicated that the victims of staff sexual misconduct or harassment during 2006 were often transferred to another facility (31%)or placed in administrative segregation or protective custody (25%). Inmate victims had been transferred to another facility in 48% of the incidents in local jails and in 19% of the incidents in State and Federal prisons. Victims were less likely to have been moved to administrative segregation or protective custody when the incident occurred in a jail(19%) than in a prison (28%). In most incidents of staff sexual misconduct or harassment (76%), victims received no medical followup, counseling or mental health treatment. Victims were given a medical examination in 6% of the incidents in prisons and jails. They were provided counseling or mental health treatment in 12% of the incidents. Methodology Measures of sexual violence In 2004 BJS developed uniform definitions of sexual violence. All incidents of inmate-on-inmate sexual violence involve sexual contacts with any person without his or her consent, or with a person who is unable to consent or refuse. The most serious incidents, nonconsensual sexual acts, include: * Contact between the penis and the vagina or the penis and the anus including penetration, however slight; or * Contact between the mouth and the penis, vagina, or anus; or * Penetration of the anal or genital opening of another person by a hand, finger, or other object. The less serious incidents, abusive sexual contacts, include: * Intentional touching, either directly or through the clothing, of the genitalia, anus, groin, breast, inner thigh, or buttocks of any person. * Incidents in which the intention is to sexually exploit (rather than to only harm or debilitate). Incidents of staff-with-inmate sexual violence are separated into two categories. Staff sexual misconduct includes any behavior or act of a sexual nature, either consensual or nonconsensual, directed toward an inmate by an employee, volunteer, official visitor, or agency representative. Such acts include: * Intentional touching of the genitalia, anus, groin, breast, inner thigh, or buttocks with the intent to abuse, arouse, or gratify sexual desire; or * Completed, attempted, threatened, or requested sexual acts; or * Occurrences of indecent exposure, invasion of privacy, or staff voyeurism for sexual gratification. Staff sexual harassment involves repeated verbal statements or comments of a sexual nature to an inmate by an employee, volunteer, official visitor, or agency representative. Such statements include demeaning references to gender or derogatory comments about body or clothing; or profane or obscene language or gestures. Since BJS first developed these definitions, correctional authorities have significantly enhanced their abilities to report uniform data on sexual violence. Authorities in 42 State and Federal prison systems were able to report incidents of abusive sexual contacts separately, as defined in the 2006 survey. Only two States limited reports to substantiated incidents. Most prison systems (44) were able to report data on staff sexual misconduct using survey definitions. Five systems were unable to separate sexual harassment from misconduct. One system did not record staff sexual harassment in a central database. Jail authorities were less likely than prison authorities to meet survey definitions. More than a third of the jail jurisdictions were unable to separate abusive sexual contacts from the more serious nonconsensual sexual acts; a quarter were unable to report staff sexual harassment separately from staff sexual misconduct. Sampling procedures The 2006 Survey of Sexual Violence was based on seven separate samples, corresponding to the different facilities covered under the Act. The following samples were drawn: The survey included all 50 State adult prison systems and the Federal Bureau of Prisons. Prison administrators were directed to report only on incidents of sexual violence that occurred within publicly operated adult facilities. A sample of 41 privately operated prison facilities was drawn to produce a 10% sample of the 408 private prisons identified in the 2005 Census of State and Federal Adult Correctional Facilities. Facilities were ranked by average daily population (ADP) in the 12-month period ending June 30, 2005. Four facilities with an ADP of more than 2,005 inmates were selected with certainty due to their size. The remaining facilities were sorted by region and ADP within region and then sampled systematically with a probability proportional to their size. A sample of 350 publicly operated jail facilities was selected based on data reported in the 2005 Census of Jail Inmates. Jail jurisdictions were sorted into five strata based on ADP during the 12-month period ending June 30, 2005, and sampled systematically to produce a representative national sample. In 2006 the second-largest jail jurisdiction in each State was selected from stratum 1 to avoid selecting with certainty the same jurisdiction as in 2005. (In previous surveys the 46 largest jails in each State were selected.) An additional 61 jail jurisdictions (forming stratum 2)were selected with certainty due to size(an ADP of 1,500 inmates or more). The remaining 2,813 jail jurisdictions in the 2005 census were then grouped into 3 strata: stratum 3 contained 1,599 jails with an ADP of 85 inmates or fewer; stratum 4 included 785 jails with an ADP of 86 to 183 inmates; and stratum 5 included 429 jails with an ADP of 284 to 1,499 inmates. Jail jurisdictions in these 3 strata were selected systematically with probabilities proportionate to their size, resulting in 46 selections from stratum 3, 52 from stratum 4, and 145 from stratum 5. Of the 350 selected jail jurisdictions, 5 did not respond to the survey: Rockdale County, GA Beaver County, PA Lycoming County, PA Greenville County, SC Ellis County, TX One of the facilities closed in 2006: Crossville City, AL 4. A sample of 5 privately operated jails was also selected based on data reported in the 2005 Census of Jail Inmates. The 42 private facilities were sorted by region, State, and ADP. Facilities were systematically sampled with probabilities proportionate to size. Three additional samples of other correctional facilities were drawn to represent: a) jails in Indian Country (10 facilities holding adult and juvenile inmates were selected from a total of 68 based on the ADP during 2004); b) military-operated facilities (all 59 facilities operated by the Armed Services in the continental U.S.); c) 14 facilities operated by Immigration and Customs Enforcement. Of the facilities selected, one had closed (Navajo Department of Corrections -- Chinle, AZ). Data for each correctional system and sampled facility are displayed in Appendix tables 1a -- 4b. In each table a measure of population size has been provided as a basis for comparison. National estimates and accuracy Survey responses were weighted to produce national estimates by type of correctional facility. Data from the Federal Bureau of Prisons, all State systems, military facilities, and ICE facilities received a weight of 1.00, since these systems and facilities were all selected (sampled with certainty). Among publicly operated jails, private prisons, private jails and jails in Indian country, facilities were assigned a weight equal to the inverse of their probability of selection. Estimates for public jail jurisdictions were adjusted for non- response by multiplying each estimate by the ratio of the total ADP in all jurisdictions to the ADP among participating jurisdictions. Survey estimates for public jail jurisdictions, private prisons and jails, and jails in Indian country are subject to sampling error. The error, as measured by an estimated sampling error, varies by the size of the estimate and the size of the base population. Estimates of the standard errors for selected survey items are presented in table 13. These standard errors may be used to construct confidence intervals around survey estimates (e.g., numbers, rates, and percentages), as well as differences in these estimates. For example, the 95% confidence interval around the number of allegations of sexual violence is approximately 6,528 plus or minus 1.96 times 143 (or 6,248 to 6,808). Detail on substantiated incidents The 2006 Survey of Sexual Violence recorded 704 substantiated incidents of sexual violence; that is, incidents that were investigated and determined to have occurred. Taking into account sampling of local jail jurisdictions, private prisons and jails, and jails in Indian country, the estimated total for the Nation was 967. Correctional authorities provided detail on 99% of all substantiated incidents. Through use of a separate incident form, the survey collected details on circumstances surrounding each incident, characteristics of victims and perpetrators, type of pressure or physical force, sanctions imposed, and victim assistance. These data are displayed in Appendix tables 5 -- 8. In response to concerns raised about the 2005 survey, BJS changed the survey item related to staff sexual misconduct and harassment. Item 27 of the 2006 survey read as follows: What was the nature of the incident? (Mark all that apply.) Pressure or abuse of power resulting in a nonconsensual sexual act Indecent exposure, invasion of privacy, or voyeurism for sexual gratification Unwanted touching for sexual gratification Sexual harassment or repeated verbal statements of a sexual nature by staff Sexual relationship between inmate and staff that appeared to be willing Other -- Specify Level of coercion unknown --------------------------------------------- The Bureau of Justice Statistics is the statistical agency of the U.S. Department of Justice. Jeffrey L. Sedgwick is the director. Allen J. Beck, Paige M. Harrison and Devon B. Adams, wrote this report. Carolyn Williams produced and edited the report, and Jayne Robinson prepared the report for publication, under the supervision of Doris J. James. Pamela H. Butler, Greta B. Clark, and Shannon Clerkin carried out data collection and processing, under the supervision of Charlene M. Sebold, Governments Division, Census Bureau, U.S. Department of Commerce. Pearl E. Chase, Patricia D. Torryson, Kathryn DiMeglio, and Lisa A. McNelis assisted in the data collection. Suzanne M. Dorinski drew the facility samples and provided sampling weights. August 2007, NCJ 218914. ---------------------------------------------- ---------------------------------------------- This report in portable document format (includes 8 appendix tables) and in ASCII and its related statistical data are available at the BJS World Wide Web Internet site: ----------------------------------------------- End of file 08/09/07 ih